About
I am an AI/ML Specialist Solutions Architect at Amazon Web Services (AWS), where I have…
Articles by Shikhar
Contributions
Activity
-
Join me at the Gartner IT Symposium in Orlando next week. Looking foward to an exciting week and seeing lots of old friends!
Join me at the Gartner IT Symposium in Orlando next week. Looking foward to an exciting week and seeing lots of old friends!
Liked by Shikhar Kwatra
-
📣 Exciting news! The MegaBeam-Mistral-7B-300K language model recently got released! 🎉 This fine-tuned model can process longer contexts of up to…
📣 Exciting news! The MegaBeam-Mistral-7B-300K language model recently got released! 🎉 This fine-tuned model can process longer contexts of up to…
Shared by Shikhar Kwatra
-
I just became a US citizen! 🇺🇸 It took 3,473 days! To be here in Silicon Valley, working in Tech, was my childhood dream. When we flew into…
I just became a US citizen! 🇺🇸 It took 3,473 days! To be here in Silicon Valley, working in Tech, was my childhood dream. When we flew into…
Liked by Shikhar Kwatra
Experience
Education
Volunteer Experience
-
Organizer at NSIT
Innocent World Charitable Society
-
Program Committee Chair
North Carolina Technical Experts Community
-
Invention Development Team - Bronze Level
Invention Development Team - Bronze Level
IBM
Licenses & Certifications
Publications
-
Model Operationalization at Edge Devices
Springer Nature
Kwatra, S., Mangla, U., Srivatsa, M. (2023). Model Operationalization at Edge Devices. In: Srivatsa, M., Abdelzaher, T., He, T. (eds) Artificial Intelligence for Edge Computing. Springer, Cham. https://doi.org/10.1007/978-3-031-40787-1_12
-
Brainvent: Bring Ideas to Life
ContentWREP Publications
The objective of this book is intended to invoke critical thinking on some practical everyday problems with an abstract yet ambiguous starting point!
Even problems that may at first seem bizarre, silly, or trivial, if approached via critical thinking, may indeed conjure mind-bending solutions that‘d intrigue amateurs as well as experts.
So put on your thinking hats and prepare to be amazed by your mind. Also, there may be multiple solutions targeting the same problem. -
ML Model Change Detection and Versioning Service
2021 IEEE International Conference on Smart Data Services (SMDS)
Many machine learning applications call for methods to recommend when to update a deployed model if the underlying distribution of incoming data drifts over time. While necessary to generate accurate predictions towards intended business outcomes, updating a model in the form of re-training and maintaining model versions is a cost intensive operation. Current approaches in change detection and versioning do not cover mechanisms for automatically detecting significant change based on feature…
Many machine learning applications call for methods to recommend when to update a deployed model if the underlying distribution of incoming data drifts over time. While necessary to generate accurate predictions towards intended business outcomes, updating a model in the form of re-training and maintaining model versions is a cost intensive operation. Current approaches in change detection and versioning do not cover mechanisms for automatically detecting significant change based on feature importance, correlation and semantic revisions occurring over time.We explore a modular approach towards model versioning focused on use-cases in the IT services industry. Our method attempts to detect and quantify changes across new and existing data sets based on statistical as well as semantic feature comparison. We demonstrate utility of our approach by implementing a change detection and model versioning service, and leveraging it for a risk analytic model built for a global IT service provider.
-
Cognitive Advisory Agent
2021 IEEE International Conference on Smart Data Services (SMDS)
Machine learning workflows involve extracting data from a dataset, training the model, predicting future behavior, and finally saving the results. At times, merging features from similar datasets or appending additional features to existing dataset may help improve trained model performance and accuracy. However, merging data is a complex operation, requiring deep analysis on features to be merged. While the state-of-the-art methods work to resolve the merging of datasets, there is a need for a…
Machine learning workflows involve extracting data from a dataset, training the model, predicting future behavior, and finally saving the results. At times, merging features from similar datasets or appending additional features to existing dataset may help improve trained model performance and accuracy. However, merging data is a complex operation, requiring deep analysis on features to be merged. While the state-of-the-art methods work to resolve the merging of datasets, there is a need for a cognitive system to performs this task efficiently.In this paper, we propose a novel cognitive advisory agent, that dynamically provides actionable insights while merging datasets to improve the performance of the model. The agent uses reward function while performing reinforcement learning in order to analyze appending features from different datasets. We demonstrate the efficacy of our method on a real-world use case and share the encouraging results.
-
AI Governance: Gain Control Over the AI Lifecycle
Linux Foundation
Artificial intelligence systems have become increasingly prevalent in everyday life and enterprise settings, and they’re now often being used to support human decision-making.
When we understand how a technology works and we can assess that it’s safe and reliable, we’re far more inclined to trust it. But even when we don’t understand the technology (do you understand how a modern automobile works?), if it has been tested and certified by a respectable body, we are inclined to trust it…Artificial intelligence systems have become increasingly prevalent in everyday life and enterprise settings, and they’re now often being used to support human decision-making.
When we understand how a technology works and we can assess that it’s safe and reliable, we’re far more inclined to trust it. But even when we don’t understand the technology (do you understand how a modern automobile works?), if it has been tested and certified by a respectable body, we are inclined to trust it. Many AI systems today are black boxes, where data is fed in and results come out. To trust a decision made by an algorithm, we need to know that it is fair, that it’s reliable and can be accounted for, and that it will cause no harm. We need assurances that AI cannot be tampered with and that the system itself is secure. We need to be able to look inside AI systems, to understand the rationale behind the algorithmic outcome, and even ask it questions as to how it came to its decision. -
What You Cannot Miss in Any AI Implementation: Fairness
Linux foundation
Artificial Intelligence (AI) is becoming a key cog in how the world works and how it lives. But the reality is that AI is not as widespread in critical enterprise workflows as it could be because it is not perceived to be safe, reliable, fair, and trustworthy. With increasing regulation, concern about brand reputation, burgeoning complexity, and a renewed focus on social justice, companies are not ready and willing to deploy a “science experiment” at scale in their operations. As Thomas J…
Artificial Intelligence (AI) is becoming a key cog in how the world works and how it lives. But the reality is that AI is not as widespread in critical enterprise workflows as it could be because it is not perceived to be safe, reliable, fair, and trustworthy. With increasing regulation, concern about brand reputation, burgeoning complexity, and a renewed focus on social justice, companies are not ready and willing to deploy a “science experiment” at scale in their operations. As Thomas J. Watson, Sr., an early chief executive of IBM said, “The toughest thing about the power of trust is that it’s very difficult to build.”
We’ve seen many newsworthy examples of AI producing unfair outcomes: blacks being discriminated against in criminal recidivism, low-income students systematically having low “predicted” exam scores when the coronavirus pandemic cancelled the real exam, men and women having different lending decisions despite having exactly the same assets, and many more. Why is this happening and what can we do about it? -
Explainable AI: A Recent Outlook and Research Study
IBM AoT
Artificial Intelligence (AI) represents ahugecommercial opportunity withthe potential to contribute $15.7 trillion to the global economy by 2030. Yet, Explainable AIis emerging as a key challenge.Explainable AI explores how an AI machine learning model produces its results and defines the most important parameters to be examined in training a neural network. Itsobjective is to produce explanations that can build trust and confidence in AI solutions with business leaders so that investments and…
Artificial Intelligence (AI) represents ahugecommercial opportunity withthe potential to contribute $15.7 trillion to the global economy by 2030. Yet, Explainable AIis emerging as a key challenge.Explainable AI explores how an AI machine learning model produces its results and defines the most important parameters to be examined in training a neural network. Itsobjective is to produce explanations that can build trust and confidence in AI solutions with business leaders so that investments and adoption of AI solutions continue to grow. IBM®has leading software and platform capabilities, world-class AI-skilled services organizations, and experience in embedding AI in enterprise business processes and solutions. This uniquely positions IBM as the partner of choice to help clients on their AI-powered digital transformation journey. In addition to IBM’s contributions to the open source community such as the AI Explainability (AIX 360) toolkit, and commercial software offerings (such as Watson OpenScale®), IBM needs to continue focusing on promoting and enabling all AI practitioners and clients on Explainability. This is a critical component of AI model development, deployment and monitoring.
-
Human-in-the-Loop Business Modelling for Emergent External Factors
IEEE HMData 2020
In the face of emergent external factors (e.g., sup-ply chain disruptions or public health crises like COVID-19),businesses must adapt their business model quickly in order to ensure service continuity. However, providing recommendations regarding changes should be made to the business model is a challenging problem. First, it requires details of interactions between different components of the business (e.g., service offerings, inventory…
In the face of emergent external factors (e.g., sup-ply chain disruptions or public health crises like COVID-19),businesses must adapt their business model quickly in order to ensure service continuity. However, providing recommendations regarding changes should be made to the business model is a challenging problem. First, it requires details of interactions between different components of the business (e.g., service offerings, inventory, staffing, demand) to understand what possible courses of action will have the most business impact. Second,automated models may provide recommendations on changes required in the business operations. However, with lack of human insight, it will be hard to verify the feasibility of these recommendations. Third, a generic model may not be able to provide good recommendations for diverse set of business models.Fourth, the model may not have enough features or training data to provide good recommendations. In this paper, we propose a novel approach to provide actionable items that can be recommended to business users given their business features and recommendations given to businesses in similar domain. Here we first use clustering to find the business domain and similar feature set of the domain. Then, we build a machine-learning model with explainable insights to provide recommendations on different business actions that can be taken to ensure business operations in the face of emergent external factors. Next we augment our approach with human-in-the-loop to improve its performance. Finally, we federate the machine-learning model in a similar domain to add more explainable and trusted insights and recommendations by other businesses. We describe our method, illustrate its utility with results from our implementation, and discuss areas for future work. (https://humanmachinedata.org/)
-
210 technical disclosures published on IP.com
IBM
Made novel contributions on 101 technical disclosures pertaining to the field of AI, cognitive, Blockchain, IoT and Hardware-software interfaces that have been published by IBM on IP.com.
-
Characterization and applications of Coherer Based Memristor
HiPEAC International Conference, 2014
Abstract : In the present paper, we study the electrical properties of cats whisker, subset of larger class of devices called coherer, which were recently proposed to be a canonical implementation of memristor. The device’s memristive properties was found to be dependent on the contact area and pressure and the state-variable controlling the device is found to be maximum current flown through the device. By controlling the state-variable, we found that the device can be programmed into multiple…
Abstract : In the present paper, we study the electrical properties of cats whisker, subset of larger class of devices called coherer, which were recently proposed to be a canonical implementation of memristor. The device’s memristive properties was found to be dependent on the contact area and pressure and the state-variable controlling the device is found to be maximum current flown through the device. By controlling the state-variable, we found that the device can be programmed into multiple resistance states and can be reset to higher resistance state, which has immediate application in programmable analog circuits. We have demonstrated that by using programming
current wave, we can achieve programming of properties such as threshold voltage, amplifier’s gain, duty cycle.Other authors -
Characterization of Discrete Novel Memristor
27th International Conference on VLSI Design and 13th International Conference on Embedded Systems, 2014
The paper discusses about a characterization board used to test the current activated memristor. The paper introduces the properties of a new memristor and its various modes and details of characterization board along with the methodology used to characterize memristor’s I-V behavior. Experimental results using various different plug and play kits for characterizing newly discovered memristor are also presented. The paper also discusses the role of such simple, easy to use and
economical…The paper discusses about a characterization board used to test the current activated memristor. The paper introduces the properties of a new memristor and its various modes and details of characterization board along with the methodology used to characterize memristor’s I-V behavior. Experimental results using various different plug and play kits for characterizing newly discovered memristor are also presented. The paper also discusses the role of such simple, easy to use and
economical boards to characterize not just the memristor but also any two-terminal device.Other authors
Patents
-
Notification content message via artificial intelligence voice response system
Filed United States
Aspects of the present invention disclose a method to derive optimal notification content to be delivered to one or plurality of users based on congregating contextual information from interconnected devices. The method includes one or more processors identifying an interaction of a user with a computing device. The method further includes determining a first set of conditions of an operating environment that includes the interaction of the user with the computing device. The method further…
Aspects of the present invention disclose a method to derive optimal notification content to be delivered to one or plurality of users based on congregating contextual information from interconnected devices. The method includes one or more processors identifying an interaction of a user with a computing device. The method further includes determining a first set of conditions of an operating environment that includes the interaction of the user with the computing device. The method further includes determining a relationship between the first set of conditions of the operating environment and the interaction of the user with the computing device. The method further includes generating a knowledge base that includes the determined relationship, the first set of conditions of the operating environment, and the interaction of the user with the computing device. The method further includes generating a notification message for the user based on the knowledge base.
-
Scalable and dynamic transfer learning mechanism
Filed United States
A computer-implemented method for adapting a deep learning model to a local environment includes collecting training data; training a common deep learning model using the collected training data; customizing the deep learning model based on characteristics specific to one of a plurality of local devices utilizing transfer learning; and deploying the customized deep learning model to the one of the plurality of local devices. The customized deep learning model may be trained across a…
A computer-implemented method for adapting a deep learning model to a local environment includes collecting training data; training a common deep learning model using the collected training data; customizing the deep learning model based on characteristics specific to one of a plurality of local devices utilizing transfer learning; and deploying the customized deep learning model to the one of the plurality of local devices. The customized deep learning model may be trained across a hierarchy comprising: a central common model trained on data from all devices in the plurality of devices; a location based model trained on a first set of data from devices in the plurality of devices within a location proximity; and a device based model trained on a data from one device in the plurality of devices. -
Relative position management of autonomous vehicles based on data bandwith requirements
Filed United States
Aspects of the present invention disclose a method for routing one or more autonomous vehicles to minimize a density of autonomous vehicles and passengers passing through network areas with oversubscribed bandwidth. The method includes one or more processors determining a bandwidth requirement of a first autonomous vehicle. The method further includes determining respective bandwidth requirement for one or more additional autonomous vehicles utilizing a wireless network. The method further…
Aspects of the present invention disclose a method for routing one or more autonomous vehicles to minimize a density of autonomous vehicles and passengers passing through network areas with oversubscribed bandwidth. The method includes one or more processors determining a bandwidth requirement of a first autonomous vehicle. The method further includes determining respective bandwidth requirement for one or more additional autonomous vehicles utilizing a wireless network. The method further includes determining a total bandwidth capacity of one or more nodes of the wireless network. The method further includes determining routing instructions from a current location of the first autonomous vehicle to a destination of the first autonomous vehicle based at least in part on the bandwidth requirement of the first autonomous vehicle and the total bandwidth capacity of the one or more nodes of the wireless network.
-
Semi-supervised reinforcement learning
Filed United States
Provided is a method, a system, and a program product for determining a policy using semi-supervised reinforcement learning. The method includes observing a state of an environment by a learning agent. The method also includes taking an action by the learning agent. The method further includes observing a new state of the environment and calculating a reward for the action taken by the learning agent. The method also includes determining whether a policy related to the learning agent should be…
Provided is a method, a system, and a program product for determining a policy using semi-supervised reinforcement learning. The method includes observing a state of an environment by a learning agent. The method also includes taking an action by the learning agent. The method further includes observing a new state of the environment and calculating a reward for the action taken by the learning agent. The method also includes determining whether a policy related to the learning agent should be changed. The determination is conducted by a teaching agent that inputs the state of the environment and the reward as features. The method can also include changing the policy related to the learning agent upon a determination that a label outputted by the teaching agent exceeds a reward threshold.
-
Optimized assignment of multiple conveyor belts
Issued United States 10773900
A computer-implemented method for assigning one or more conveyor belts to convey a plurality of items. The computer-implemented method determines an availability of one or more conveyor belts, determines a quantity of a plurality of items to be loaded onto the one or more conveyor belts, and assigns the one or more conveyor belts to convey the plurality of items, based on the determined quantity of the plurality of items to be loaded onto the one or more conveyor belts. The computer-implemented…
A computer-implemented method for assigning one or more conveyor belts to convey a plurality of items. The computer-implemented method determines an availability of one or more conveyor belts, determines a quantity of a plurality of items to be loaded onto the one or more conveyor belts, and assigns the one or more conveyor belts to convey the plurality of items, based on the determined quantity of the plurality of items to be loaded onto the one or more conveyor belts. The computer-implemented method further tracks a location of each of the plurality of items on the assigned one or more conveyor belts, associates each of the plurality of items with a user, and directs the user to the one or more conveyor belts based on the tracked location of each of the plurality of items associated with the user.
-
MOBILE THERMAL BALANCING OF DATA CENTERS
Filed United States
A tool for mobile thermal load balancing in a data center . The tool determines a heat signature pattern for one or more data center units based on a plurality of heat signatures over a pre - determined time period . The tool determines a high heat signature zone within a confined region of the data center based , at least in part , on a current arrangement of the one or more data center units . Responsive to a determination that a high heat signature zone within the data center exceeds a…
A tool for mobile thermal load balancing in a data center . The tool determines a heat signature pattern for one or more data center units based on a plurality of heat signatures over a pre - determined time period . The tool determines a high heat signature zone within a confined region of the data center based , at least in part , on a current arrangement of the one or more data center units . Responsive to a determination that a high heat signature zone within the data center exceeds a predetermined heat threshold , the tool rearranges the one or more data center units within the high heat signature zone to a region within the data center that balances a thermal load associated with the high heat signature zone across the data center.
-
Monitoring security threat during travel
Issued United States 10573184
An approach for notifying a passenger of a route deviation during travel. The approach includes calculating a passenger route deviation from a predicted travel route and comparing the passenger route deviation to a predefined threshold route deviation. The approach further includes providing the passenger a notification of a route deviation if the deviation exceeds a predefined route deviation threshold. In addition, the approach determines if the passenger acknowledged the route deviation…
An approach for notifying a passenger of a route deviation during travel. The approach includes calculating a passenger route deviation from a predicted travel route and comparing the passenger route deviation to a predefined threshold route deviation. The approach further includes providing the passenger a notification of a route deviation if the deviation exceeds a predefined route deviation threshold. In addition, the approach determines if the passenger acknowledged the route deviation notification within a predetermined time period and if the passenger fails to acknowledge the notification, then provides the notification to one or more third parties.
-
Methods and systems for enhancing viewer engagement with content portions
Issued United States US10547582B1
Embodiments for managing viewer engagement of online content by one or more processor are described. Content posted on an online channel is detected. A viewer engagement component associated with the content is received. A portion of the content is selected based on the received viewer engagement component. A signal representative of the selected portion of the content is generated.
-
Augmented reality based driver assistance
Issued United States 10469769
A head mounted display (HMD) device worn by a user in a vehicular computing environment (VCE) can be augmented with video by measuring, within the VCE, a sight angle between a user's directional focus and a vehicle's directional orientation and augmenting, in response to a determination that the measured sight angle exceeds a threshold limit, an interface of the HMD device with a video feed of continuous live content captured within a central viewing area.
-
Automated management of undesired code use based on predicted valuation and risk analysis
Filed United States
Automatically managing ameliorative process that controls undesired code use in an application based on predicted valuation and risk vectors. A multidimensional risk vectors may be received in relation to current and future use of undesired code. Amelioration actions may be classifies by using a custom trained machine learning model. An optimal amelioration action may be selected based on at least one optimization objective function. The selected amelioration action may be taken to control…
Automatically managing ameliorative process that controls undesired code use in an application based on predicted valuation and risk vectors. A multidimensional risk vectors may be received in relation to current and future use of undesired code. Amelioration actions may be classifies by using a custom trained machine learning model. An optimal amelioration action may be selected based on at least one optimization objective function. The selected amelioration action may be taken to control undesired code use in application. Examples of an amelioration action may include changing of the characteristics of a user interface and stopping an application or a component of the application from running.
-
Contextual user profile photo selection
Filed United States
Selecting and displaying a profile photo may include determining viewer characteristics of a viewer who is triggering an action to open an image file that shows a photo image. A context associated with viewing of the profile photo and photo image characteristics of a plurality of photos may be determined. Based on the viewer characteristics, the photo image characteristics and the context associated with viewing of the profile photo, a photo from the plurality of photos may be selected…
Selecting and displaying a profile photo may include determining viewer characteristics of a viewer who is triggering an action to open an image file that shows a photo image. A context associated with viewing of the profile photo and photo image characteristics of a plurality of photos may be determined. Based on the viewer characteristics, the photo image characteristics and the context associated with viewing of the profile photo, a photo from the plurality of photos may be selected. Responsive to determining that the selected photo is to be transformed, a transformed version of the photo image may be generated and displayed.
-
PROTECTING CONFIDENTIAL CONVERSATIONS ON DEVICES
Issued United States 10382620
A first mobile device may receive user-input data for generation of an electronic message. The electronic message may be generated as the user-input data is received. The first mobile device may identify sensitive information, within the electronic message. The first mobile device may secure the sensitive information. The first mobile device may disable a screen capture function of the first mobile device. The first mobile device may send the electronic message to a second mobile device. The…
A first mobile device may receive user-input data for generation of an electronic message. The electronic message may be generated as the user-input data is received. The first mobile device may identify sensitive information, within the electronic message. The first mobile device may secure the sensitive information. The first mobile device may disable a screen capture function of the first mobile device. The first mobile device may send the electronic message to a second mobile device. The sending of the electronic message to the second mobile device may include the sending of the secured sensitive information.
-
Suggesting content for an electronic document based on a user's cognitive context
Filed United States
Embodiments of the invention are directed to a computer-implemented method that includes receiving, by a controller, information of a user. The information is targeted for inclusion within an application document. The method also includes determining a cognitive context of the user. The method also includes generating, by the controller, an output data comprising a suggested content. The suggested content is based at least in part on the determined cognitive context.
-
METHODS AND SYSTEMS FOR MANAGING MOBILE DEVICES WITH REFERENCE POINTS
Issued United States 10375619
Embodiments for managing mobile devices by one or more processors are described. A plurality of mobile devices are detected at a first location. After the plurality of mobile devices are detected at the first location, a first of the plurality of mobile devices is detected traveling to a second location while a second of the plurality of mobile devices remains at the first location. In response to the plurality of mobile devices being detected at the first location and the first of the…
Embodiments for managing mobile devices by one or more processors are described. A plurality of mobile devices are detected at a first location. After the plurality of mobile devices are detected at the first location, a first of the plurality of mobile devices is detected traveling to a second location while a second of the plurality of mobile devices remains at the first location. In response to the plurality of mobile devices being detected at the first location and the first of the plurality of mobile devices being detected traveling to the second location while the second of the plurality of mobile devices remains at the first location, data detected by the second of the plurality of mobile devices at the first location is enabled to be transmitted to the first of the plurality of mobile devices at the second location.
-
COGNITIVE STATE AWARE ACCELERATED ACTIVITY COMPLETION AND AMELIORATION
Filed United States
Methods and systems for cognitive state aware accelerated activity completion and amelioration are disclosed . A method includes : collecting data related to a cognitive state of a user and a physical state of the user ; determining a type of an activity performed by the user and an amount of work for the activity ; determining the cognitive state of the user and the physical state of the user based on the data related to the cognitive state of the user and the physical state of the user…
Methods and systems for cognitive state aware accelerated activity completion and amelioration are disclosed . A method includes : collecting data related to a cognitive state of a user and a physical state of the user ; determining a type of an activity performed by the user and an amount of work for the activity ; determining the cognitive state of the user and the physical state of the user based on the data related to the cognitive state of the user and the physical state of the user , determining a predicted completion time for the activ ity based on the type of the activity , the amount of work for the activity , the cognitive state of the user , and the physical state of the user ; and displaying a progress indicator includ ing a portion of the activity that has been completed and an estimated completion time.
-
Autonomous emergency evacuation
Issued United States 10796582
Aspects of the present disclosure relate to emergency evacuation. An emergency can be detected at an emergency location. A type of the emergency can be determined. Prospective safe locations proximate to the emergency location can be identified. A safety rating of each prospective safe location can be determined based on the type of emergency. A number of at-risk individuals at the emergency location can be determined. A subset of drop-off locations of the prospective safe locations that have a…
Aspects of the present disclosure relate to emergency evacuation. An emergency can be detected at an emergency location. A type of the emergency can be determined. Prospective safe locations proximate to the emergency location can be identified. A safety rating of each prospective safe location can be determined based on the type of emergency. A number of at-risk individuals at the emergency location can be determined. A subset of drop-off locations of the prospective safe locations that have a safety rating that satisfies a safety threshold can be selected, the subset of drop-off locations satisfying a size limit required for the number of at-risk individuals. A set of autonomous vehicles required for the number of at-risk individuals can then be determined. The set of autonomous vehicles can be deployed to the emergency location.
-
GENERATING USER INTERFACE PREVIEWS
Filed United States
A computer - implemented method includes: determining, by a computer device, an interaction pattern of a website based on actions of plural users at the website ; generating, by the computer device, a demonstration video based on the inter action pattern; and presenting, by the computer device, the demonstration video to a user visiting the website for a first time .
-
DEVICE PROTECTION BASED ON PREDICTION AND CONTEXTUAL ANALYSIS
Filed United States
A user's cognitive state of a user who is using a device is estimated . Based on past history of use of the device and the estimated user's cognitive state , a possible deleterious user action on the device is detected . Based on the detected possible deleterious user action on the device , the device can be caused to perform an amelioration action for time period P.
-
Indoor environment personalization preferences
Filed United States
A method, computer system, and a computer program product for environment personalization is provided. The present invention may include initializing a profile of a user. The present invention may include defining a baseline within the profile of the user. The present invention may include tracking a plurality of user data. The present invention may include storing the tracked plurality of user data in a tracked user database. The present invention may lastly include optimizing an environmental…
A method, computer system, and a computer program product for environment personalization is provided. The present invention may include initializing a profile of a user. The present invention may include defining a baseline within the profile of the user. The present invention may include tracking a plurality of user data. The present invention may include storing the tracked plurality of user data in a tracked user database. The present invention may lastly include optimizing an environmental condition based on the tracked plurality of user data.
-
INTELLIGENT ACTION RECOMMENDATION
Issued United States 10318639
Intelligent action recommendation in a conversation monitors content of a conversation among at least two participants in real time. An overall goal of the conversation that represents a motivation for at least one of the participants to engage in the conversation is identified. The overall goal of the conversation is to be achieved upon termination of the conversation. At least one conversational input relevant to the monitored content and having a likelihood of advancing the conversation to…
Intelligent action recommendation in a conversation monitors content of a conversation among at least two participants in real time. An overall goal of the conversation that represents a motivation for at least one of the participants to engage in the conversation is identified. The overall goal of the conversation is to be achieved upon termination of the conversation. At least one conversational input relevant to the monitored content and having a likelihood of advancing the conversation to successful achievement of the overall goal of the conversation is identified. The identified conversational input is communicated to at least one of the participants in the conversation.
-
OPTIMAL CONTENT IDENTIFICATION FOR LEARNING PATHS
Filed United States
A method selects content based on a learning relevancy targeted to specific recipients. One or more processors extract semantic features, which provide meanings of concepts, from each content asset from a plurality of content assets. The processor(s) utilize a clustering algorithm to group entries in the content assets based on the semantic features in order to form hierarchical consolidated entries for the semantic features, where each hierarchical consolidated entry is associated with one of…
A method selects content based on a learning relevancy targeted to specific recipients. One or more processors extract semantic features, which provide meanings of concepts, from each content asset from a plurality of content assets. The processor(s) utilize a clustering algorithm to group entries in the content assets based on the semantic features in order to form hierarchical consolidated entries for the semantic features, where each hierarchical consolidated entry is associated with one of the semantic features. The processor(s) then provide a representation of each of the hierarchical consolidated entries based on a target audience criteria.
-
CONTEXTUAL API CAPTCHA
Filed United States
A request for use of an application programming interface( API ) is received. Context associated with the request is determined. Based on the context, a challenge is generated, which can be used for determining whether to permit the use of the API. A response to the challenge is received. Based on the response, the request can be facilitated for using the API
-
Enabling attention by leveraging a user-effective communication channel
Filed United States
An exemplary apparatus includes a memory embodying computer executable instructions and at least one processor, coupled to the memory, and operative by the computer executable instructions to facilitate a method. The method includes tracking a topic and a context of an electronic conversation in a first communication channel by using a machine learning method to process the electronic conversation; identifying confusion in a participant of the electronic conversation regarding a change in the…
An exemplary apparatus includes a memory embodying computer executable instructions and at least one processor, coupled to the memory, and operative by the computer executable instructions to facilitate a method. The method includes tracking a topic and a context of an electronic conversation in a first communication channel by using a machine learning method to process the electronic conversation; identifying confusion in a participant of the electronic conversation regarding a change in the topic or the context; identifying a user-effective communication channel for alerting the participant to the change in the topic or the context; and alerting the participant of the change in the topic or the context via the user-effective communication channel.
-
Implementing enhanced devops process for cognitive search solutions
Filed United States
A method and system are provided for implementing enhanced cognitive searches optimized to integrate deployment with development testing. An initial body of works is ingested into a system capable of answering questions. A series of queries is analyzed against the initial body of works utilizing a set of scorers utilizing criteria to form assessments, wherein each scorer uses the criteria against the query and the initial body of works to form the assessment. The assessments are analyzed to…
A method and system are provided for implementing enhanced cognitive searches optimized to integrate deployment with development testing. An initial body of works is ingested into a system capable of answering questions. A series of queries is analyzed against the initial body of works utilizing a set of scorers utilizing criteria to form assessments, wherein each scorer uses the criteria against the query and the initial body of works to form the assessment. The assessments are analyzed to determine a usefulness of a set of entries in the initial body of works. Content are deployed using the identified selected first set of entries as deployment ready, enabling enhanced cognitive search results.
-
Form-based transactional conversation system design
Filed United States
A method, computer system, and computer program product for form-based conversation system design are provided. The embodiment may include ingesting, by a processor, a plurality of forms from a given domain. The embodiment may also include extracting indicators of required input fields from the ingested plurality of forms. The embodiment may further include generating a required input list based on the extracted indicators of the required input fields to update a size of the required input…
A method, computer system, and computer program product for form-based conversation system design are provided. The embodiment may include ingesting, by a processor, a plurality of forms from a given domain. The embodiment may also include extracting indicators of required input fields from the ingested plurality of forms. The embodiment may further include generating a required input list based on the extracted indicators of the required input fields to update a size of the required input list. The embodiment may also include determining transactional intents based on the required input list. The embodiments may further include generating a dialog flow that satisfies the determined transactional intents.
-
METHODS AND SYSTEMS FOR MANAGING MULTIPLE RECIPIENT ELECTRONIC COMMUNICATIONS
Filed United States
Embodiments for managing electronic communications by one or more processors are described. A plurality of historical communications are analyzed. Each of the plurality of historical communications is associated with at least one of a plurality of recipients. Reactions of the plurality of recipients to each of the plurality of historical communications associated therewith are analyzed. The analysis of the reactions includes normalizing the reactions to determine a consensus of the reactions. A…
Embodiments for managing electronic communications by one or more processors are described. A plurality of historical communications are analyzed. Each of the plurality of historical communications is associated with at least one of a plurality of recipients. Reactions of the plurality of recipients to each of the plurality of historical communications associated therewith are analyzed. The analysis of the reactions includes normalizing the reactions to determine a consensus of the reactions. A current communication associated with a sender and the plurality of recipients is analyzed. The current communication is modified based on the analyses of the plurality of historical communications, the reactions of the plurality of recipients, and the current communication. The modified current communication is caused to be transmitted to at least one of the sender and at least some of the plurality of recipients.
-
Videoconferencing dynamic host controller
Filed United States
Aspects of the present invention generate a vector value representing a state of a host device hosting a videoconference that is attended by a plurality of participant videoconference devices, wherein the host device controls audio and video distributed for the videoconference to others of the participant videoconference devices, evaluate the vector value to determine a change in the vector value during the videoconference, identify a plurality of second host devices of the plurality of…
Aspects of the present invention generate a vector value representing a state of a host device hosting a videoconference that is attended by a plurality of participant videoconference devices, wherein the host device controls audio and video distributed for the videoconference to others of the participant videoconference devices, evaluate the vector value to determine a change in the vector value during the videoconference, identify a plurality of second host devices of the plurality of participant videoconference devices that are each enabled to control the audio and video distributed for the videoconference to others of the participant videoconference devices, and in response to determining that the change in the vector value exceeds a threshold, switch control of the videoconference from the host device to one of the second host devices.
-
EVENT DETECTION AND NOTIFICATION SYSTEM
Filed United States
Embodiments of the present invention provide a method, system, and program product for event detection and/or notification of events occurring. A user monitoring system sends messages based on determinations of impacts associated with the user. The monitoring system receives data from at least one computing device affixed to a user. The data corresponds to a pattern of movement of the user. The monitoring system determines that the user is performing an activity that has an associated known…
Embodiments of the present invention provide a method, system, and program product for event detection and/or notification of events occurring. A user monitoring system sends messages based on determinations of impacts associated with the user. The monitoring system receives data from at least one computing device affixed to a user. The data corresponds to a pattern of movement of the user. The monitoring system determines that the user is performing an activity that has an associated known pattern of impacts. The system generates a customized pattern of impacts for the activity by modifying the known pattern of impacts based on the pattern of movement of the user. The system generates a message based on a comparison of measured movement data of the user and the customized pattern of impacts.
-
PARKING CONTINUITY WITH UNUSED DURATION BETWEEN AUTOMATED VEHICLES
Filed United States
An artificial neural network trained to predict the availability of an unused duration of a parking space based on input features is executed. Input features may include at least a contextual situation associated with the second entity, a behavior factor associated with a first entity that has been using the parking space, geographical location and time, events occurring within a threshold distance from the parking space. The artificial neural network may be further trained to output a transfer…
An artificial neural network trained to predict the availability of an unused duration of a parking space based on input features is executed. Input features may include at least a contextual situation associated with the second entity, a behavior factor associated with a first entity that has been using the parking space, geographical location and time, events occurring within a threshold distance from the parking space. The artificial neural network may be further trained to output a transfer affinity based on the predicted availability of an unused duration, the contextual situation associated with the second entity and the behavior factor associated with the first entity. Based at least on the transfer affinity, the second entity can be selected. The unused duration can be transferred to the second entity from the first entity. The transferring can also include storing a payment and associated computation as a blockchain node in a blockchain.
-
DATABASE ENDORSEMENT ARCHITECTURE
Filed United States
An example operation may include one or more of computing historical patterns related to fraudulent attempts from a transaction log, predicting future fraud attempts from public data, correlating the historical patterns and the predicted future fraud attempts, modifying one or more first endorsement policies based on the correlations, preventing modifying one or more second endorsement policies, the one or more second endorsement policies providing read-only access, and adding the modified one…
An example operation may include one or more of computing historical patterns related to fraudulent attempts from a transaction log, predicting future fraud attempts from public data, correlating the historical patterns and the predicted future fraud attempts, modifying one or more first endorsement policies based on the correlations, preventing modifying one or more second endorsement policies, the one or more second endorsement policies providing read-only access, and adding the modified one or more first endorsement policies to a smart contract.
-
AUTO-EVOLVING DATABASE ENDORSEMENT POLICIES
Filed United States
An example operation may include one or more of computing historical patterns related to fraudulent attempts from a transaction log, predicting future fraud attempts from public data, correlating the historical patterns and the predicted future fraud attempts, modifying one or more endorsement policies based on the correlations, and adding the modified one or more endorsement policies to a smart contract.
-
METHODS AND SYSTEMS FOR REMOTE SENSING DEVICE CONTROL BASED ON FACIAL INFORMATION
Filed United States
Embodiments for controlling a remote sensing device by one or more processors are described. Facial information associated with a plurality of organisms is received. A remote sensing direction for a remote sensing device is selected based on the received facial information. A signal representative of the remote sensing direction is generated.
-
Method and system for configuring a virtual reality environment
Filed United States
A method for creating a customized virtual shopping experience may include receiving physical-world shopping data of a first user. The physical-world shopping data may be collected while the first user is physically present in a physical-world shopping environment. The method may also include configuring a virtual shopping environment based at least in part on the physical-world shopping data of the first user, and displaying to the first user the virtual shopping environment.
-
INTELLIGENT CLUSTER LEARNING IN AN INTERNET OF THINGS (IOT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for intelligent cluster learning in an Internet of Things (IoT) computing environment by a processor. One or more interactive actions, engagements, behaviors, communications, biometric data, parameters, or combination thereof relating to one or more IoT devices may be analyzed. A user profile may be clustered into a cluster of profiles according to the one or more interactive actions, engagements, behaviors, communications, biometric data, parameters, or combination thereof using…
Embodiments for intelligent cluster learning in an Internet of Things (IoT) computing environment by a processor. One or more interactive actions, engagements, behaviors, communications, biometric data, parameters, or combination thereof relating to one or more IoT devices may be analyzed. A user profile may be clustered into a cluster of profiles according to the one or more interactive actions, engagements, behaviors, communications, biometric data, parameters, or combination thereof using one or more machine learning operations. Access to one or more functional assets and parameters associated with the one or more IoT devices may be automatically enabled or disabled.
-
Dynamic communication session filtering
Filed United States
Provided is a system, method, and computer program product for filtering spoken content in a preferred format to a plurality of users within a communication session A processor may detect spoken content from a first user and a second user of a plurality of users within a communication session over a network. Speech features may be extracted from the detected spoken content. The first user and second user may be identified based on the extracted speech features. The spoken content of the first…
Provided is a system, method, and computer program product for filtering spoken content in a preferred format to a plurality of users within a communication session A processor may detect spoken content from a first user and a second user of a plurality of users within a communication session over a network. Speech features may be extracted from the detected spoken content. The first user and second user may be identified based on the extracted speech features. The spoken content of the first user and the second user may be filtered according to preferences determined from profiles of the plurality of users. The spoken content of the first user and the second user may be transmitted to the plurality of users in a preferred format based on preferences from the profiles.
-
In-vehicle adjustment to destination environmental conditions
Filed United States
A method, computer system, and a computer program product for adjusting a plurality of environmental conditions of a vehicle based on a plurality of environmental conditions of an expected destination is provided. The present invention may include monitoring the plurality of environmental conditions of a vehicle by utilizing at least one vehicle device sensor. The present invention may then include retrieving the plurality of environmental conditions of an expected destination of the vehicle…
A method, computer system, and a computer program product for adjusting a plurality of environmental conditions of a vehicle based on a plurality of environmental conditions of an expected destination is provided. The present invention may include monitoring the plurality of environmental conditions of a vehicle by utilizing at least one vehicle device sensor. The present invention may then include retrieving the plurality of environmental conditions of an expected destination of the vehicle. The present invention may also include determining a delta between the monitored plurality of environmental conditions of the vehicle and the retrieved plurality of environmental conditions of the expected destination of the vehicle. The present invention may further include adjusting, over a gradual period of time, the monitored plurality of environmental conditions of the vehicle.
-
DYNAMIC INTERACTION BEHAVIOR COMMENTARY
Filed United States
A method, computer system, and computer program product for dynamic interaction behavior commentary are provided. The embodiment may include receiving, by a processor, a plurality of posted comments. The embodiment may also include analyzing the plurality of posted comments based on a semantic distance of the topic of discussion measured by content analysis of the plurality of posted comments. The embodiment may further include determining optimal frameworks for a plurality of users in response…
A method, computer system, and computer program product for dynamic interaction behavior commentary are provided. The embodiment may include receiving, by a processor, a plurality of posted comments. The embodiment may also include analyzing the plurality of posted comments based on a semantic distance of the topic of discussion measured by content analysis of the plurality of posted comments. The embodiment may further include determining optimal frameworks for a plurality of users in response to a comment from a user within the plurality of posted comments. The embodiment may also include identifying interaction behavior responses containing statistically significant words or phrases that may improve a tone of the next comment when the user responds to a certain post. The embodiment may further include notifying the user when the next statement is generated based on a pre-identified or dynamically-learned interaction behavior comment. The embodiment may also include prompting the user, via a graphical user interface, with suggested timelines and places where a comment should be inserted.
-
CONTEXT ENABLED SENDER COMMUNICATION AWARENESS ALERT
Filed United States
A context-driven sender communication awareness method, system, and computer program product include detecting an intent of a sender sending an electronic communication to a receiver over a communication channel, establishing a potential risk to the receiver in connection with receiving the electronic communication on a device, determining an estimated time duration in which the established potential risk is applicable, and alerting the sender about the potential risk that results from…
A context-driven sender communication awareness method, system, and computer program product include detecting an intent of a sender sending an electronic communication to a receiver over a communication channel, establishing a potential risk to the receiver in connection with receiving the electronic communication on a device, determining an estimated time duration in which the established potential risk is applicable, and alerting the sender about the potential risk that results from delivering of the electronic communication within the estimated time duration.
-
Dynamic Natural Language Processing
Filed United States
Embodiments relate to an intelligent computer platform to support natural language (NL) processing. The request is analyzed and a lexical answer type (LAT) related to the received request is identified. A knowledge graph (KG) related to the LAT is identified and leveraged to extract a first concept related to the LAT and a second concept related to the first concept. First and second cluster are created, with the first cluster having the LAT and first concept as qualifiers, and the second…
Embodiments relate to an intelligent computer platform to support natural language (NL) processing. The request is analyzed and a lexical answer type (LAT) related to the received request is identified. A knowledge graph (KG) related to the LAT is identified and leveraged to extract a first concept related to the LAT and a second concept related to the first concept. First and second cluster are created, with the first cluster having the LAT and first concept as qualifiers, and the second cluster having the first and second concepts as qualifiers. Each of the formed clusters is populated with one or more documents. An inter-cluster assessment is conducted based on the relevancy of the populated document(s) to the received input. In addition, a machine learning model (MLM) corresponding to the KG is identified and utilized to selectively augment the MLM with the LAT, first and second concepts, and a corresponding relationship to the inter-cluster assessment.
-
Dynamic Query Processing and Document Retrieval
Filed United States
Embodiments relate to an intelligent computer platform to receive a request for processing against a corpus. The request is analyzed and a lexical answer type (LAT), a first concept relevant to the received request and a second concept related to the identified first concept, are each identified. The LAT, together with the first and second concepts are utilized to create a first and second cluster. Documents are selectively populated into the clusters based on the respective LAT and concept…
Embodiments relate to an intelligent computer platform to receive a request for processing against a corpus. The request is analyzed and a lexical answer type (LAT), a first concept relevant to the received request and a second concept related to the identified first concept, are each identified. The LAT, together with the first and second concepts are utilized to create a first and second cluster. Documents are selectively populated into the clusters based on the respective LAT and concept qualifiers. The clusters are subject to sorting based on relevancy to the received request.
-
COGNITIVE COMMUNICATION CHANNEL-ADAPTATION BASED ON CONTEXT
Filed United States
Methods, computer program products, and systems are presented. The methods include, for instance: predicting an upcoming communication content in a current channel based on learning input data from a plurality of sources relevant to a communication activity of a user. A user response to the predicted content is assessed and a communication channel model for the factors including the input data is selected. A user satisfaction level with the current channel is checked against a predefined…
Methods, computer program products, and systems are presented. The methods include, for instance: predicting an upcoming communication content in a current channel based on learning input data from a plurality of sources relevant to a communication activity of a user. A user response to the predicted content is assessed and a communication channel model for the factors including the input data is selected. A user satisfaction level with the current channel is checked against a predefined threshold for the selected communication channel model. For the current channel that is less than satisfactory, an adaptive action is taken to adapt the current channel to another for the user.
-
METHODS AND SYSTEMS FOR AUTO-FILLING FIELDS OF ELECTRONIC DOCUMENTS
Filed United States
Embodiments for managing an electronic document by one or more processors are described. An entry for a first of a plurality of fillable fields of an electronic document is received. A recommended entry for at least a second of the plurality of fillable fields is determined based on at least one data source associated with a user. A signal representative of the determined recommended entry for the at least a second of the plurality of fillable fields is generated.
-
METHODS AND SYSTEMS FOR MANAGING CHATBOTS BASED ON TOPIC SENSITIVITY
Filed United States
Embodiments for managing a chatbot by one or more processors are described. A communication is received from a first individual. The presence of a second individual within a proximity of a speaker is detected. A response to the communication is determined based on the communication and the detected presence of the second individual. The determined response is caused to be executed.
-
METHODS AND SYSTEMS FOR MANAGING CONTENT BASED ON PROXY UNDERSTANDING
Filed United States
Embodiments for managing communications by one or more processors are described. A communication sent to a first individual is detected. A reaction score for the first individual consuming the communication is calculated based on at least one data source associated with at least one of the first individual and a second individual. The communication is caused to be rendered with an indication of the calculated reaction score by a computing node associated with the first individual.
-
METHODS AND SYSTEMS FOR MANAGING MULTI-CHANNEL COMPUTING ENVIRONMENTS TO OPTIMIZE USER EFFICIENCY
Filed United States
Embodiments for managing a multi-channel computing environment by one or more processors are described. A user simultaneously interacting with a plurality of computing nodes is detected. A computing node prioritization strategy for future user interaction with at least one of the plurality of computing nodes is determined based on at least one data source associated with at least one of the detected user interaction with the plurality of computing nodes and the user. The future user interaction…
Embodiments for managing a multi-channel computing environment by one or more processors are described. A user simultaneously interacting with a plurality of computing nodes is detected. A computing node prioritization strategy for future user interaction with at least one of the plurality of computing nodes is determined based on at least one data source associated with at least one of the detected user interaction with the plurality of computing nodes and the user. The future user interaction with at least one of the plurality of computing nodes is at least limited based on the determined computing node prioritization strategy.
-
METHODS AND SYSTEMS FOR MANAGING COMMUNICATIONS AND RESPONSES THERETO
Filed United States
Embodiments for managing communications by one or more processors are described. Communication content is received from an individual. At least one communication channel to which to send the communication content is selected. The at least one communication channel is selected from a plurality of communication channels based on the communication content and at least one data source associated with the plurality of communication channels. A notification of the at least one selected communication…
Embodiments for managing communications by one or more processors are described. Communication content is received from an individual. At least one communication channel to which to send the communication content is selected. The at least one communication channel is selected from a plurality of communication channels based on the communication content and at least one data source associated with the plurality of communication channels. A notification of the at least one selected communication channel is generated.
-
METHODS AND SYSTEMS FOR MANAGING VOICE COMMANDS AND THE EXECUTION THEREOF
Filed United States
Embodiments for managing voice commands by one or more processors are described. The receiving of a voice command from an individual is detected. An action associated with the voice command is caused to be at least temporarily prevented from being executed based on at least one data source associated with the individual.
-
OPTIMAL DRIVING CHARACTERISTIC ADJUSTMENT FOR AUTONOMOUS VEHICLES
Filed United States
Embodiments for implementing intelligent driving comfort adjustment of an autonomous vehicle by a processor. A user experience satisfaction level may be determined during a journey within an autonomous vehicle according to historical user experience satisfaction levels, a user profile, one or more contextual factors, or a combination thereof. One or more performance characteristics of the autonomous vehicle may be adjusted if a user experience satisfaction level is less than a predetermined…
Embodiments for implementing intelligent driving comfort adjustment of an autonomous vehicle by a processor. A user experience satisfaction level may be determined during a journey within an autonomous vehicle according to historical user experience satisfaction levels, a user profile, one or more contextual factors, or a combination thereof. One or more performance characteristics of the autonomous vehicle may be adjusted if a user experience satisfaction level is less than a predetermined threshold.
-
METHODS AND SYSTEMS FOR MANAGING VOICE RESPONSE SYSTEMS BASED ON REFERENCES TO PREVIOUS RESPONSES
Filed United States
Embodiments for managing a voice response system by one or more processors are described. At least one communication from an individual is received. A reference to a previous response executed by a voice response system is identified within the at least one received communication. A response to the at least one received communication is determined based on the previous response executed by the voice response system.
-
METHODS AND SYSTEMS FOR MANAGING HAZARD RISK BASED ON LOCATION AND INCIDENT DATA
Filed United States
Embodiments for managing hazard risk by one or more processors are described. A presence of an individual at a location is detected. A hazard risk score is calculated based on at least one data source associated with at least one of the individual and the location. A notification of the calculated hazard risk score is caused to be generated.
-
Personalized video and memories creation based on enriched images
Filed United States
A method for creating content includes making a corpus of images available, adding a plurality of tag data to each of the images, receiving a query, extracting a trigger from the query, identifying a set of the images in the corpus, wherein identified images have tag data matching the extracted trigger, and creating a video comprising the identified images.
-
DERIVING CONFIDENCE SCORES BASED ON DEVICE SHARING
Filed United States
Systems and methods for deriving confidence scores based on device sharing are disclosed. In embodiments, a method includes receiving, by a computing device, sharing event data from a remote computing device in a comparative confidence environment, the sharing event data including usage data regarding the sharing of an electronic device between a first participant and a second participant obtained by the remote computing device during a sharing event; calculating, by the computing device, a…
Systems and methods for deriving confidence scores based on device sharing are disclosed. In embodiments, a method includes receiving, by a computing device, sharing event data from a remote computing device in a comparative confidence environment, the sharing event data including usage data regarding the sharing of an electronic device between a first participant and a second participant obtained by the remote computing device during a sharing event; calculating, by the computing device, a comparative confidence score for the first participant and the second participant based on the sharing event data; and enabling, by the computing device, the first participant to gain access to a resource of the second participant based on the comparative confidence score.
-
Simplifying electronic communication based on dynamically structured contact entries
Filed United States
A first set of data signals is received from a first user device associated with a first user. Temporal properties associated with the first user are determined from the first set of data signals. The temporal properties may include a current cognitive state of the first user and environment context associated the first user. Responsive to detecting the temporal properties, contact entries associated with the first user device are dynamically structured. A second user is selected from the…
A first set of data signals is received from a first user device associated with a first user. Temporal properties associated with the first user are determined from the first set of data signals. The temporal properties may include a current cognitive state of the first user and environment context associated the first user. Responsive to detecting the temporal properties, contact entries associated with the first user device are dynamically structured. A second user is selected from the dynamically structured contact entries. The first user device may be automatically or autonomously triggered to initiate a communication with the second user via the first user device.
-
Personalized table of contents (toc) generation based on search results
Filed United States
The present invention is a system and method that generates a Table of Contents (TOC) customized to the user knowledge about the concept(s) in the user query and the specific context and preferences of the user. The invention identifies search concepts within the search queries, receives search results, and splits the search results into one or more result segments. In a preferred embodiment, a correlation strength between concepts in one or more of the result segments and the user search…
The present invention is a system and method that generates a Table of Contents (TOC) customized to the user knowledge about the concept(s) in the user query and the specific context and preferences of the user. The invention identifies search concepts within the search queries, receives search results, and splits the search results into one or more result segments. In a preferred embodiment, a correlation strength between concepts in one or more of the result segments and the user search query, along with reference to the user knowledge, context, and/or preferences determine which result segments are selected in sequenced to form the TOC.
-
Methods and systems for managing content storage
Filed United States
Embodiments for managing content by one or more processors are described. A content portion associated with a user is detected. A storage scheme for the content portion is selected based on at least one data source associated with utilization of previous content portions by the user. The content portion is caused to be stored on a computing device according to the selected storage scheme. -
DETECTING AND REDUCING BIAS IN MACHINE LEARNING MODELS
Filed United States
A method identifies and removes bias from a machine learning model. A user/computer inputs a plurality of input training data into a machine learning system to generate an output of labeled output data. The user/computer evaluates the labeled output data according to a consistency metric to associate the labeled output data with a corresponding consistency assessment. The user/computer selects each labeled output data having a consistency assessment indicating a consistency assessment that is…
A method identifies and removes bias from a machine learning model. A user/computer inputs a plurality of input training data into a machine learning system to generate an output of labeled output data. The user/computer evaluates the labeled output data according to a consistency metric to associate the labeled output data with a corresponding consistency assessment. The user/computer selects each labeled output data having a consistency assessment indicating a consistency assessment that is greater than a predetermined threshold to form a labeled output data subset, and then creates additional labeling for the labeled output data subset. The user/computer utilizes the additional labeling to distinguish each labeled training data from labeled output data subset as being mislabeled and biased, and then adjusts the learning machine based on the labeled output data subset being mislabeled and biased.
-
FILTERING MEDIA DATA IN AN INTERNET OF THINGS (IOT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for intelligent media data filtering in an Internet of Things (IoT) computing environment by a processor. User preferences for the media data may be determined according to a plurality of identified contextual factors. At least a portion of the media data may be filtered according to one or more user classifications.
-
DEVICE-TO-DEVICE WIRELESS POWER TRANSFER TO PROVIDE CONTENT DISPLAY CONTINUITY
Filed United States
A system transfers power between a plurality of devices. Content is displayed to a group of users. An amount of power required by the system to at least complete display of the content is determined. In response to the required amount of power exceeding the power in a rechargeable power source of the system, one or more devices that are associated with users of the group are identified, wherein the identified devices each include a power source. Power is wirelessly received from the identified…
A system transfers power between a plurality of devices. Content is displayed to a group of users. An amount of power required by the system to at least complete display of the content is determined. In response to the required amount of power exceeding the power in a rechargeable power source of the system, one or more devices that are associated with users of the group are identified, wherein the identified devices each include a power source. Power is wirelessly received from the identified one or more devices. Embodiments of the present invention further include a method and program product for transferring power between a plurality of devices in substantially the same manner described above.
-
SOCIAL MEDIA CONTROL PROVISIONING BASED ON A TRUSTED NETWORK
Filed United States
A social media control method, system, and computer program product, include detecting a contextual meaning of a content of a third-party post on a user's social media account, classifying the content into user-defined categories based on an overlap between the contextual meaning and the user-defined categories, identifying an undesired piece of information from the content, identifying a trusted online friend within a social media circle of the user, and providing a one-time access control to…
A social media control method, system, and computer program product, include detecting a contextual meaning of a content of a third-party post on a user's social media account, classifying the content into user-defined categories based on an overlap between the contextual meaning and the user-defined categories, identifying an undesired piece of information from the content, identifying a trusted online friend within a social media circle of the user, and providing a one-time access control to the identified trusted online friend to rectify the undesired content in the third-party post.
-
Cognitive Hierarchical Content Distribution
Filed United States
Embodiments relate to a system, program product, and method for use with an intelligent computer platform and cognitive processing and associated distribution. The embodiments support a mechanism for dynamically sharing critical and non-critical information responsive to a classification model, such that only relevant information or part of the information is shared. As the classification model is subject to modification, the dynamic sharing mechanism is dynamically updated to reflect such…
Embodiments relate to a system, program product, and method for use with an intelligent computer platform and cognitive processing and associated distribution. The embodiments support a mechanism for dynamically sharing critical and non-critical information responsive to a classification model, such that only relevant information or part of the information is shared. As the classification model is subject to modification, the dynamic sharing mechanism is dynamically updated to reflect such modification. Similarly, as an associated document or file is subject to modification, dynamic processing of the document or file takes place responsive to the classification model. The dynamic classification and document processing employ NLP and ML models to support the associated functionality.
-
Methods and systems for personalized rendering of presentation content
Filed United States
Embodiments for rendering content, such as a teleconference, by one or more processors are described. A personalized presentation rendering profile for at least one individual is received. Portions of the content are selected based on the personalized presentation rendering profile. A rendering of the content is caused to be generated. The rendering of the content includes rendering the selected portions of the content in a manner different than other portions of the content.
-
Heat-based pattern recognition and event determination for adaptive surveillance control in a surveillance system
Filed United States
A method for heat-based control of a surveillance system is provided. The method may include detecting a first pattern of heat transferred from a heat source based on a first sensor dataset corresponding to a first event, determining an expected pattern of heat to be transferred from the heat source during a second event based on the first pattern of heat transfer, and generating a surveillance model based on the expected pattern of heat transfer. The method may further include detecting a…
A method for heat-based control of a surveillance system is provided. The method may include detecting a first pattern of heat transferred from a heat source based on a first sensor dataset corresponding to a first event, determining an expected pattern of heat to be transferred from the heat source during a second event based on the first pattern of heat transfer, and generating a surveillance model based on the expected pattern of heat transfer. The method may further include detecting a second pattern of heat transferred from the heat source based on a second sensor dataset corresponding to the second event, detecting the heat source during the second event with respect to the second pattern of heat transfer, and determining a threat level corresponding to a security risk posed by the heat source with respect to the environment.
-
Maintaining voice conversation continuity
Filed United States
An inventive system includes a first electronic device, such as a cellular phone, that implements a method of detecting a voice conversation via the first electronic device; detecting an interruption of continuity of the voice conversation; and adapting operation of the first electronic device to maintain the continuity of the voice conversation. For example, adapting operation of the first electronic device may include auto-completing the voice conversation during the interruption of…
An inventive system includes a first electronic device, such as a cellular phone, that implements a method of detecting a voice conversation via the first electronic device; detecting an interruption of continuity of the voice conversation; and adapting operation of the first electronic device to maintain the continuity of the voice conversation. For example, adapting operation of the first electronic device may include auto-completing the voice conversation during the interruption of continuity. As another example, adapting operation of the first electronic device may include generating a text summary of the voice conversation during the interruption of continuity by activating a speech-to-text function and then transmitting to a second electronic device an electronic signal encoding the text summary. The method may include detecting a level of importance of the voice conversation by identifying a topic of the voice conversation.
-
METHODS AND SYSTEMS FOR MANAGING MEDICAL ANOMALIES
Filed United States
Embodiments for managing medical anomalies by one or more processors are described. Medical data associated with an individual is received. The medical data is detected by a first computing device. The first computing device includes a mobile electronic device. The received medical data is compared to stored medical data associated with the individual. Based on the comparison of the received data to the stored data, a control signal for a second computing device is caused to be generated.
-
Cognitive location and navigation services for custom applications
Filed United States
According to various embodiments, systems, computer program products, and computer implemented methods for cognitive location and navigation services for custom applications are disclosed. More specifically, the cognitive location and navigation services include, but are not limited to: multilingual, point-specific communication; cognitive navigational guidance through a tourist attraction; and beacon-based communication with a vehicle are disclosed. For instance, one embodiment of…
According to various embodiments, systems, computer program products, and computer implemented methods for cognitive location and navigation services for custom applications are disclosed. More specifically, the cognitive location and navigation services include, but are not limited to: multilingual, point-specific communication; cognitive navigational guidance through a tourist attraction; and beacon-based communication with a vehicle are disclosed. For instance, one embodiment of multilingual, point-specific communication includes receiving a request for delivery of announcements relating to a particular event; receiving or retrieving itinerary information corresponding to the particular event and a user that submitted the request; receiving audio data from a location corresponding to the particular event based at least in part on the itinerary information; translating the audio data into a plurality of translated messages, each translated message being in a different language; and providing at least one of the translated messages to the user that submitted the request.
-
DYNAMIC MODIFICATION OF MEDIA CONTENT IN AN INTERNET OF THINGS (IoT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for optimizing media content in an Internet of Things (IoT) environment. User information is collected by a first, embedded IoT computing device associated with a user. The collected user information is transmitted from the first computing device to a second, content-displaying IoT computing device. Characteristics of media content rendered by the second computing device are then rendered according to an analyzation of the user information received from the first computing device.
-
METHODS AND SYSTEMS FOR MANAGING COMMUNICATION SESSIONS FOR DISCUSSION COMPLETENESS
Filed United States
Embodiments for managing a communication session by one or more processors are described. A communication session including a plurality of communications sent between a plurality of entities is monitored. A recommended communication for the communication session is determined based on the plurality of communications and at least one data source associated with at least one of the plurality of entities. A signal representative of the recommended communication is generated.
-
METHODS AND SYSTEMS FOR MANAGING MULTIPLE TOPIC ELECTRONIC COMMUNICATIONS
Filed United States
Embodiments for managing electronic communications by one or more processors are described. A plurality of electronic messages are received. Each of the plurality of electronic messages is associated with a respective one of a plurality of topics. A portion of the plurality of topics is selected. A signal representative of the plurality of electronic messages that are associated with the selected portion of the plurality of topics is generated.
-
INTELLIGENT FEEDBACK AND CONTEXT DRIVEN WEB NAVIGATION
Filed United States
Embodiments for implementing intelligent web navigation by at least a portion of a processor. One or more uniform resource locators (URLs) may be accessed (e.g., opened) according a user history of Internet activities, user preferences, one or more current user activities, calendar data, or a combination thereof. Selected feedback may be gathered (e.g., collected) from the user relating to the one or more URLs according to activities of the user performed on the one or more URLs. The one or…
Embodiments for implementing intelligent web navigation by at least a portion of a processor. One or more uniform resource locators (URLs) may be accessed (e.g., opened) according a user history of Internet activities, user preferences, one or more current user activities, calendar data, or a combination thereof. Selected feedback may be gathered (e.g., collected) from the user relating to the one or more URLs according to activities of the user performed on the one or more URLs. The one or more URLs may be classified into one of a plurality of classifications to indicate a degree of relevancy to the user.
-
METHODS AND SYSTEMS FOR MANAGING CONTENT ON MULTIPLE DEVICES
Filed United States
Embodiments for managing content by one or more processors are described. A rendering of a first content portion by a first computing node is detected. A second content portion is received. The second content portion is associated with the first content portion. A second computing node is selected based on at least one of a location of the second computing node, an activity of an individual, and the second content portion. The second content portion is caused to be rendered by the second…
Embodiments for managing content by one or more processors are described. A rendering of a first content portion by a first computing node is detected. A second content portion is received. The second content portion is associated with the first content portion. A second computing node is selected based on at least one of a location of the second computing node, an activity of an individual, and the second content portion. The second content portion is caused to be rendered by the second computing node.
-
ACTIVE VEHICLE VIRTUAL REALITY PREVENTION OF PREDICTIVE MOTION SICKNESS
Filed United States
A negator module of a predictive motion system determines initial parameters for a passenger profile using a virtual reality system of an autonomous vehicle. The negator module receives upcoming driving conditions from an autonomous navigation system of the autonomous vehicle during a ride in which the passenger resides in a seat of the autonomous vehicle and uses the virtual reality system. Using a cognitive model, the negator module predicts a cognitive state of the passenger based on the…
A negator module of a predictive motion system determines initial parameters for a passenger profile using a virtual reality system of an autonomous vehicle. The negator module receives upcoming driving conditions from an autonomous navigation system of the autonomous vehicle during a ride in which the passenger resides in a seat of the autonomous vehicle and uses the virtual reality system. Using a cognitive model, the negator module predicts a cognitive state of the passenger based on the passenger profile and the upcoming driving conditions. The negator module determines commands for actuators coupled to the seat and commands for the virtual reality system that match the predicted cognitive state of the passenger. The negator module sends the commands to the actuators and the virtual reality system to be executed.
-
INTELLIGENT HEARING AID
Filed United States
Embodiments for using an intelligent hearing aid device by a processor. Audio data may be received and analyzed for a user according to a plurality of user preferences and interests, historical activity patterns of the user, or a combination thereof. One or more hearing assistive actions may be performed in relation to the audio data to facilitate hearing according to the plurality of user preferences and interests, historical activity patterns of the user, or a combination thereof.
-
INTELLIGENT UPDATING OF MEDIA DATA IN A COMPUTING ENVIRONMENT
Filed United States
Embodiments for intelligent updating of media data by a processor. One or more data sources may be searched to identify updated data content for streamed media data. The data content may be updated in the streamed media data with the updated data content from the one or more data sources according to a defined priority.
-
FACILITATION OF COGNITIVE CONFLICT RESOLUTION BETWEEN PARTIES
Filed United States
Embodiments for facilitating cognitive conflict resolution between parties by a processor. An appropriateness of communications, behavior, actions or events associated with one or more users may be cognitively interpreted according to a plurality of identified contextual factors during a conflict resolution. One or more corrective actions may be suggested to mitigate a possible negative impact of the communications, behavior, actions or events upon the one or more users if the interpreted…
Embodiments for facilitating cognitive conflict resolution between parties by a processor. An appropriateness of communications, behavior, actions or events associated with one or more users may be cognitively interpreted according to a plurality of identified contextual factors during a conflict resolution. One or more corrective actions may be suggested to mitigate a possible negative impact of the communications, behavior, actions or events upon the one or more users if the interpreted appropriateness is less than a predetermined threshold.
-
INTELLIGENT TELECONFERENCE OPERATIONS IN AN INTERNET OF THINGS (IOT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for intelligent teleconference operations in an Internet of Things (IoT) computing environment by a processor. A communication connection for a conference call session may be cognitively initiated or terminated with one or more users according to one or more parameters associated with a user profile, a schedule of the one or more users, activities of daily living (“ADL”), one or more contextual factors, or a combination thereof.
-
METHODS AND SYSTEMS FOR MANAGING TEXT IN RENDERED IMAGES
Filed United States
Embodiments for managing text in rendered images by one or more processors are described. An image rendered by a display device is detected. The rendered image includes an object component and a text component at least partially overlapping the object component. A size of the object component of the image is caused to change. Based on a comparison of the changed size of the object component of the image to a size of the text component of the image, one or more text modification option is caused…
Embodiments for managing text in rendered images by one or more processors are described. An image rendered by a display device is detected. The rendered image includes an object component and a text component at least partially overlapping the object component. A size of the object component of the image is caused to change. Based on a comparison of the changed size of the object component of the image to a size of the text component of the image, one or more text modification option is caused to be rendered by the display device. An indication of a selection of at least one of the one or more text modification option is received. The text component of the image is caused to change based on the selected one or more text modification option.
-
INTELLIGENT ILLUMINATION AND SOUND CONTROL IN AN INTERNET OF THINGS (IOT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for implementing intelligent illumination and sound characteristics control in an Internet of Things (IoT) computing environment by a processor. An intensity and quality of an illumination and sound characteristics of displayed media content in an IoT computing environment may be monitored and analyzed. A risk factor of one or more users may be estimated for both current settings and a predicted adjustment of the illumination and sound characteristics associated with displayed media…
Embodiments for implementing intelligent illumination and sound characteristics control in an Internet of Things (IoT) computing environment by a processor. An intensity and quality of an illumination and sound characteristics of displayed media content in an IoT computing environment may be monitored and analyzed. A risk factor of one or more users may be estimated for both current settings and a predicted adjustment of the illumination and sound characteristics associated with displayed media content according to the analyzing. The illumination, the sound characteristics, or combination thereof of the displayed media content may be adjusted upon determining the risk factor for the one or more users is above a selected threshold.
-
SOCIAL MEDIA/NETWORK ENABLED DIGITAL LEARNING ENVIRONMENT WITH ATOMIC REFACTORING
Filed United States
A set of learner data associated with a learner is received. The set of learner data includes at least a real-time data stream, e.g., derived from instrumentation, in a social network enabled digital learning environment. Based on the set of learner data, at least one learning improvement strategy is determined for improving the learner's learning capability. Atomic level mentorship activities are generated. Each atomic level mentorship activity is targeted with a specific skill and knowledge…
A set of learner data associated with a learner is received. The set of learner data includes at least a real-time data stream, e.g., derived from instrumentation, in a social network enabled digital learning environment. Based on the set of learner data, at least one learning improvement strategy is determined for improving the learner's learning capability. Atomic level mentorship activities are generated. Each atomic level mentorship activity is targeted with a specific skill and knowledge item and having an estimated duration period. At least one mentorship is identified from at least one social network platform. An initial incentive structure comprising at least an amount of payment and payment modality is created. The atomic level mentorship activities with the initial incentive structure are presented.
-
PROVIDING LOCATION-BASED SERVICES USING GEO-FENCING TRACKING TECHNIQUES
Filed United States
A computer system tracks geographical locations traversed by a user to provide location based services. One or more geographical locations traversed by a user are tracked via a portable location device. Geo-fencing areas corresponding to the one or more geographical areas are determined. One or more geo-fencing area profile scores are generated for the user based on attributes of the traversal of the geo-fencing areas. Re-location of the user to a new geographic location is detected via the…
A computer system tracks geographical locations traversed by a user to provide location based services. One or more geographical locations traversed by a user are tracked via a portable location device. Geo-fencing areas corresponding to the one or more geographical areas are determined. One or more geo-fencing area profile scores are generated for the user based on attributes of the traversal of the geo-fencing areas. Re-location of the user to a new geographic location is detected via the portable location device. Location-based services are provided for the new geographic location based on at least one geo-fencing area profile score of the user. Embodiments of the present invention further include a method and program product for tracking geographical locations traversed by a user to provide location based services in substantially the same manner described above.
-
METHODS AND SYSTEMS FOR MANAGING PERSONAL DEVICE SECURITY
Filed United States
Embodiments for managing personal device security by one or more processors are described. Information associated with activity of a personal device is received. The information is detected utilizing a sensor coupled to the personal device. A signal representative of the received information is generated. An operational state of the personal device is caused to be changed after the generating of the signal.
-
INTELLIGENT NOTIFICATION MODE SWITCHING IN USER EQUIPMENT
Filed United States
Embodiments for intelligent notification mode switching in user equipment (UE) by a processor. Data may be received from one or more UEs associated with one or more users. Electronic communication switching factors associated with the one or more users may be identified. An implication of a switching notification mode on the one or more UEs based on the electronic communication switching factors may be learned. An electronic communication modes of the one or more UEs associated with the one or…
Embodiments for intelligent notification mode switching in user equipment (UE) by a processor. Data may be received from one or more UEs associated with one or more users. Electronic communication switching factors associated with the one or more users may be identified. An implication of a switching notification mode on the one or more UEs based on the electronic communication switching factors may be learned. An electronic communication modes of the one or more UEs associated with the one or more users may be cognitively switched according to the learned implication.
-
INTELLIGENT RECOMMENDATION OF GUIDANCE INSTRUCTIONS
Filed United States
Embodiments for providing intelligent recommendation of guidance instructions to a user by a processor. A user may be cognitively recognized performing a selected task on an item using one or more Internet of Things (“IoT”) computing devices. One or more guidance instructions may be provided/suggested for assisting with performing the selected task according to identified contextual factors. The user may be cognitively guided to perform the selected task using the one or more guidance…
Embodiments for providing intelligent recommendation of guidance instructions to a user by a processor. A user may be cognitively recognized performing a selected task on an item using one or more Internet of Things (“IoT”) computing devices. One or more guidance instructions may be provided/suggested for assisting with performing the selected task according to identified contextual factors. The user may be cognitively guided to perform the selected task using the one or more guidance instructions.
-
INTELLIGENT INDUCTIVE CHARGING OF MULTIPLE DEVICES BASED ON POWER PREFERENCES
Filed United States
Embodiments for inductive charging multiple devices computing environment by a processor. One or more devices may be prioritized for receiving a power charge via a wireless charging device according to one or more power charging parameters. A selected amount of the power charge may be provided to the one or more prioritized devices, via the wireless charging device, according to the one or more power charging parameters.
-
Methods and systems for managing viewer engagement of live streaming events
Filed United States
Embodiments for managing viewer engagement of live streaming events by one or more processors are described. A plurality of viewer engagement components associated with a live streaming event are received. A grade for each of the plurality of viewer engagement components is determined based on information associated with the live streaming event. A rendering of at least some of the plurality of viewer engagement components by a computing node is caused to at least be delayed based on the…
Embodiments for managing viewer engagement of live streaming events by one or more processors are described. A plurality of viewer engagement components associated with a live streaming event are received. A grade for each of the plurality of viewer engagement components is determined based on information associated with the live streaming event. A rendering of at least some of the plurality of viewer engagement components by a computing node is caused to at least be delayed based on the determined grades.
-
Intelligent design structure selection in an internet of things (iot) computing environment
Filed United States
Embodiments for intelligent design structure selection in an Internet of Things (IoT) computing environment by a processor. Levels of satisfaction and behavior patterns of one or more users having similar user profiles influencing the behavior patterns may be learned and evaluated. One or more design structure solutions may be cognitively suggested according to the levels of satisfaction and the behavior patterns.
-
Advanced web page content management
Filed United States
Controlling contents of interfaces may include receiving multiple interfaces presented on a display screen. The multiple interfaces may include at least one primary task interface and each of the multiple interfaces may present at least one content item. A user engagement level with respect to content items presented via the multiple interfaces may be estimated. A degree of user effectiveness with respect to progression toward completing a primary task presented via the primary task interface…
Controlling contents of interfaces may include receiving multiple interfaces presented on a display screen. The multiple interfaces may include at least one primary task interface and each of the multiple interfaces may present at least one content item. A user engagement level with respect to content items presented via the multiple interfaces may be estimated. A degree of user effectiveness with respect to progression toward completing a primary task presented via the primary task interface may be determined. A content item may be selected from the multiple interfaces based on the user engagement level and the degree of user effectiveness. A presentation of the selected content item may be made in a controlled manner on an unused portion of at least one of the multiple interfaces.
-
INTELLIGENT USER NOTIFICATION DURING AN EVENT IN AN INTERNET OF THINGS (IOT) COMPUTING ENVIRONMENT
Filed United States
Embodiments for intelligent notification during an event in an Internet of Things (IoT) computing environment by a processor. During a scheduled event, user activity for one or more users and geolocation of the one or more active UEs linked to the event may be monitored using a machine learning operation. A notification may be provided to one or more UEs associated with the user to alert the user to respond to the query upon determining the user failed to respond to the query during the event…
Embodiments for intelligent notification during an event in an Internet of Things (IoT) computing environment by a processor. During a scheduled event, user activity for one or more users and geolocation of the one or more active UEs linked to the event may be monitored using a machine learning operation. A notification may be provided to one or more UEs associated with the user to alert the user to respond to the query upon determining the user failed to respond to the query during the event. The notification may be terminated or disabled upon learning the user or an alternative user responded to the query.
-
SMART CONTACT LENS SYSTEM WITH COGNITIVE ANALYSIS AND AID
Filed United States
A smart lens system may include a hardware processor coupled to a smart contact lens, which may include a microprocessor, an image capturing sensor, and a wireless communication interface. The smart lens system may determine reading factors associated with a user with respect to the user reading content item, based on data related to the user's physiological movements captured by the smart contact lens. The reading factors may include a user's cognitive state determined at least from analyzing…
A smart lens system may include a hardware processor coupled to a smart contact lens, which may include a microprocessor, an image capturing sensor, and a wireless communication interface. The smart lens system may determine reading factors associated with a user with respect to the user reading content item, based on data related to the user's physiological movements captured by the smart contact lens. The reading factors may include a user's cognitive state determined at least from analyzing data representing the user's physiological movements. Responsive to determining that the user's cognitive state is above a threshold value, the smart lens system may generate an assistive action associated with the content item and transmit a signal to at least one assistive device to perform the assistive action. The assistive device performs the assistive action responsive to receiving the signal.
-
Facilitating personalized down-time activities
Filed United States
A method includes producing a user profile in response to user data; obtaining context for estimating the duration of a down-time; estimating the duration of the down-time; determining experience enhancement factors in response to the user profile and the estimated duration of the down-time; determining an activity selection model in response to the context, the user data, and the experience enhancement factors; generating personalized engaging activities by applying the activity selection…
A method includes producing a user profile in response to user data; obtaining context for estimating the duration of a down-time; estimating the duration of the down-time; determining experience enhancement factors in response to the user profile and the estimated duration of the down-time; determining an activity selection model in response to the context, the user data, and the experience enhancement factors; generating personalized engaging activities by applying the activity selection model to a plurality of potential activities; partitioning the estimated duration of the down-time into segments; sequencing the personalized engaging activities to fit into the segments; and presenting the personalized engaging activities to the user via a computing device. A computer readable medium embodies instructions for the method and an apparatus including a processor is operative to facilitate the method.
-
Facilitating micro-task performance during down-time
Filed United States
A method includes obtaining a user profile that includes a user's affinities to each of a plurality of micro-tasks; estimating a duration of down-time for the user; estimating task selection factors responsive to the user profile and to the estimated duration of down-time; generating a list of micro-tasks responsive to the task selection factors; ranking the list of micro-tasks according to the user's affinities; partitioning the down-time according to the ranked list of micro-tasks; and…
A method includes obtaining a user profile that includes a user's affinities to each of a plurality of micro-tasks; estimating a duration of down-time for the user; estimating task selection factors responsive to the user profile and to the estimated duration of down-time; generating a list of micro-tasks responsive to the task selection factors; ranking the list of micro-tasks according to the user's affinities; partitioning the down-time according to the ranked list of micro-tasks; and presenting one or more of the micro-tasks to the user, according to the partitioning of the down-time, by altering operation of a user device.
-
Performing Real-Time Analytics for Customer Care Interactions
Filed United States
A system, computer program product, and method are provided to analyze an interaction associated with a dialogue. An intelligent real-time analytics using natural language processing (NLP) monitors and analyzes customer dialogue. The system performs analytics on a detected or received dialogue to mine data associated with attributes unique to one or more human communication patterns. The NLP-based system generates and measures a tone, and classifies the tone into a category.
-
COGNITIVE RIDE SCHEDULING
Filed United States
Embodiments for facilitating ride scheduling by a processor. An occurrence of an event associated with a user may be predicted based on user data. One or more ride scheduling parameters relating to the event may be determined. One or more ride scheduling models may be determined satisfying the ride scheduling parameters. Facilitate scheduling a vehicle for the user according to the ride scheduling models.
-
ENHANCEMENT OF COMMUNICATIONS TO A USER FROM ANOTHER PARTY USING COGNITIVE TECHNIQUES
Filed United States
Embodiments for enhancing communications for a user by a processor. An appropriateness of communications may be learned for communicating with a user based on one or more disabilities relating to the user. One or more customized communications may be created based on the learned appropriateness of the communications for the user. The customized communications may be modified, adjusted, and/or polished based on reaction to the customized communications by the user, a caregiver of the user, or a…
Embodiments for enhancing communications for a user by a processor. An appropriateness of communications may be learned for communicating with a user based on one or more disabilities relating to the user. One or more customized communications may be created based on the learned appropriateness of the communications for the user. The customized communications may be modified, adjusted, and/or polished based on reaction to the customized communications by the user, a caregiver of the user, or a combination thereof.
-
EDUCATION REWARD SYSTEM AND METHOD
Filed United States
A reward learning system includes a user interface configured to receive modes of user information related to a state of the user. A cognitive computing system includes a reward system. The reward system includes a dynamically upgraded profile model of the user which is updated in accordance with the user information related to the state. The reward system is updated by machine learning employing feedback from user responses measured by the user interface and searched information by the…
A reward learning system includes a user interface configured to receive modes of user information related to a state of the user. A cognitive computing system includes a reward system. The reward system includes a dynamically upgraded profile model of the user which is updated in accordance with the user information related to the state. The reward system is updated by machine learning employing feedback from user responses measured by the user interface and searched information by the cognitive computing system. The reward system includes an increasing reward protocol based on learned user preferences and responses and rewarded in accordance with user achievements.
-
METHODS AND SYSTEMS FOR RENDERING HOLOGRAPHIC CONTENT
Filed United States
Embodiments for rendering holographic content by one or more processors are described. A security level associated with holographic content is determined. A security clearance associated with a viewer within a proximity of a holographic display is determined. The holographic content is caused to be rendered in a first mode or a second mode by the holographic display based on the determined security level associated with the holographic content and the determined security clearance associated…
Embodiments for rendering holographic content by one or more processors are described. A security level associated with holographic content is determined. A security clearance associated with a viewer within a proximity of a holographic display is determined. The holographic content is caused to be rendered in a first mode or a second mode by the holographic display based on the determined security level associated with the holographic content and the determined security clearance associated with the viewer.
-
ARTIFICIAL INTELLIGENCE PROFILING
Filed United States
Technical solutions are described for controlling an artificial intelligent gaming device. For example, a computer-implemented method includes identifying an electronic opponent profile for an opponent. The computer-implemented method also includes selecting, from a profile repository, a first set of robot profiles, where robots trained using robot profiles from the first set of robot profiles have previously defeated the opponent. The computer-implemented method also includes selecting, from…
Technical solutions are described for controlling an artificial intelligent gaming device. For example, a computer-implemented method includes identifying an electronic opponent profile for an opponent. The computer-implemented method also includes selecting, from a profile repository, a first set of robot profiles, where robots trained using robot profiles from the first set of robot profiles have previously defeated the opponent. The computer-implemented method also includes selecting, from the profile repository, a second set of robot profiles, where robots trained using robot profiles from the second set of robot profiles have previously lost to the opponent. The computer-implemented method also includes generating a current robot profile based on the first set of robot profiles and the second robot profiles. The computer-implemented method also includes configuring a robot according to the current robot profile to play against the opponent.
-
METHODS AND SYSTEMS FOR DETECTING AND RECTIFYING FAULTS IN 3D PRINTING PROCESSES
Filed United States
During a 3D printing process to form a 3D printed object, an at least partially completed 3D printed object formed by the 3D printing process is scanned. An error on the at least partially completed 3D printed object is detected based on the scanning of the at least partially completed 3D printed object. The error on the at least partially completed 3D printed object is corrected. The correcting of the error includes causing at least some material from the partially completed 3D printed object…
During a 3D printing process to form a 3D printed object, an at least partially completed 3D printed object formed by the 3D printing process is scanned. An error on the at least partially completed 3D printed object is detected based on the scanning of the at least partially completed 3D printed object. The error on the at least partially completed 3D printed object is corrected. The correcting of the error includes causing at least some material from the partially completed 3D printed object to be removed.
-
DATA-ANALYSIS-BASED VALIDATION OF PRODUCT REVIEW DATA AND LINKING TO SUPPLY CHAIN RECORD DATA
Filed US
Data-analysis-based validation of product review data and linking to product record
data are provided to facilitate forwarding product-related guidance. -
Intelligent temporary contact sharing between user equipment
Filed United States
Embodiments for intelligent temporary contact sharing between user equipment (UE) by a processor. A temporary exchange of contact data may be provided between a first UE and a second UE via a wireless communication network according to one or more policies, user preferences, historical data, a defined physical or virtual proximity between the first UE and the second UE, or a combination thereof. The contact data may be automatically deleted between the first UE and the second UE according to…
Embodiments for intelligent temporary contact sharing between user equipment (UE) by a processor. A temporary exchange of contact data may be provided between a first UE and a second UE via a wireless communication network according to one or more policies, user preferences, historical data, a defined physical or virtual proximity between the first UE and the second UE, or a combination thereof. The contact data may be automatically deleted between the first UE and the second UE according to the one or more policies, the user preferences, the historical data, the defined physical or virtual proximity between the first UE and the second UE, or a combination thereof.
Courses
-
Advanced Logic Design
CSEE 4823
-
CAD of Digital Systems
CSEE 6861
-
Computer Architecture
CSEE 4824
-
Digital VLSI Circuits
ELEN 4321
-
Embedded Systems
CSEE 4840
-
Formal Verification
COMSE 6998
-
Internet of Things
EECSE4764
-
Internet, Economics and Society
ELEN E6767
-
System On Chip
CSEE 6868
-
VLSI Design Lab
ELEN 6350
-
Advanced Robotics and Embedded Training
-
-
Analog and Digital Integrated Circuits
-
-
Biomedical Instrumentation
-
-
Computer Networking
-
-
Control Systems
-
-
Digital Signal Processing
-
-
Industrial Organisation and Management
-
-
Linear Integrated Circuits
-
-
Process Control
-
-
Transducers and Measurements
-
-
Stanford University : Machine learning
-
Projects
-
Cognitive Audio Summarizer
https://youtu.be/e8_syHKs5uk
Audio_summarizer system: Watson Text to Speech inculcated with Text summarizer for audio summarizing huge files -
Real time Mood analysis based Notifier mechanism
Created a Cascade CNN classifier integrated with OpenCV and Twilio package to send customized notifications to the user based on varying emotions.
-
Optimal Machine Anomaly prediction mechanism
-
Created Machine learning model for optimally extracting and predicting the machine failure based on correlation with 51 timestamped sensor values. Model selection and hyperparameter tuning improved the accuracy to 99.7%. Won popular choice award at Hackathon. Youtube link: https://youtu.be/TGIGfPqgMi0
-
Intelligent Infant Monitoring System
-
Developed an intelligent infant distraction mechanism in order to maximize the safety of the infant. The technology involves usage of Python Flask for processing the commands on the Server side and communication with priority based hardware distraction mechanisms, including, Huzzah based RC car and Raspberry pi based music system. Along with the automated aforementioned distraction techniques, automatic notification is also triggered to the linked devices to remotely monitor the infant in real…
Developed an intelligent infant distraction mechanism in order to maximize the safety of the infant. The technology involves usage of Python Flask for processing the commands on the Server side and communication with priority based hardware distraction mechanisms, including, Huzzah based RC car and Raspberry pi based music system. Along with the automated aforementioned distraction techniques, automatic notification is also triggered to the linked devices to remotely monitor the infant in real time and control the distraction mechanisms manually with the smartphone android application.
Other creators -
8 Bit SAR-ADC Fabrication
-
Designed an 8-bit Successive Approximation Resistor based Analog to digital Converter and performed a comparative analysis of the custom ADC chip ( using 130nm IBM technology) fabricated on a 28-pin DIP Package. Created a dual layer PCB for performing comparative analysis with other measurement parameters.
Other creatorsSee project -
CU Racing
-
CU Racing is a retro sprites graphics based racing game with two degrees of freedom for screen scrolling and realtime processing for sprite rotation. The sprites graphic engine was inspired from the TI TMS9918 Video Display Controller (VDC)1, however it was extended to provide a more modern graphics look and feel. In addition to developing the architecture for this improved sprites graphics engine, our team designed an interface which could parse png image files and load them dynamically…
CU Racing is a retro sprites graphics based racing game with two degrees of freedom for screen scrolling and realtime processing for sprite rotation. The sprites graphic engine was inspired from the TI TMS9918 Video Display Controller (VDC)1, however it was extended to provide a more modern graphics look and feel. In addition to developing the architecture for this improved sprites graphics engine, our team designed an interface which could parse png image files and load them dynamically during runtime. This dynamic pattern loading ability allowed our team to sidestep the limited number of bits available of onboard RAM on the Altera Cyclone V FPGA chip.
Other creatorsSee project -
Custom Hardware Prefetcher
-
Developed a modified Custom Hardware Offset Prefetcher mechanism derived from the basic version of next line prefetcher which was delivered in for dpc2 championship. The modified offset prefetcher would work akin to a next line prefetcher in case of a cache hit and would create a dynamically programmed offset based on cache misses.
-
Comparative Verification of Hardware Accelerators
-
-
ScrumIT- Agile Project Management Android Application
The app has been created by keeping in mind all the problems that a SCRUM master faces in managing a project. It provides simple solution in the following ways :
- Login using User name and Password.
- Ability to create multiple projects.
- Ability to add tickets(tasks) to the created projects.
- Edit the ticket details like changing priority, status, assigned member etc.
- Visualization for the entire project in the form of graphs & charts.
Other creators -
Bit-by-Bit (B3) Product
-
Worked in the three-person team for prototype development of Bit-by-Bit (B3): a product that connects everyone to design ’Internet of Things’. The work has been patented and is now supported by Microsoft Ventures in London.
Other creatorsSee project -
Development of Indigenous and affordable myo-electric arm prostheses
-
• Myo- electric arm prosthesis to help upper limb amputees in collaboration with Institute for the Physically Handicapped, New Delhi.
• Developed EMG sensor board for feature extraction of required signals and Multi-Level Perceptron algorithm for feature classification and matching for proper adaptation.
• This Project got selected in Honeywell’s Young Innovators Challenge, 2014.
Other creatorsSee project -
Memristor I-V Characterization Board Kit
Designed a low-cost DIY I-V characterization Memristor PCB kit using Arduino and Raspberry-Pi and validated various analog circuits using the same.
-
Non-invasive heart-rate measurement system
• In a team, developed the heart-rate monitor using IR sensor and Cortex M3, interfaced with graphic LCD by employing the principles of Plethysmography.
-
“Pac–Man” on Digital Storage Oscilloscope
• Designed a single player ”Pac–Man” on a Digital Storage Oscilloscope (DSO) as part of Atmel Oscilloscope Design Challenge, CEDT, NSIT, New Delhi.
• Coded the levels of the 3D Pac-man game-base. In a team, established Graphic Display in X–Y mode of DSO using two R-2R ladder networks based Digital to analog Converters and gameplay through a basic controller interfaced by Arduino.
Other creatorsSee project -
”Snake” game on an 8x8 LED Matrix
• In a 2 member team, developed the game “Snake” on LED Matrix, played through a basic controller interfaced with Arduino.
Other creatorsSee project -
Bachelor’s Project: Implementing speaker recognition using Student’s t–Mixture Model
-
-
Projects in Wireless standards
-
• Developed Arduino based Home Automation System module using two-Zigbee.
• Designed and coded Zigbee based Wireless Remote control Robotic car.
• Developed Zigbee based digital clock with an on-screen wireless data capturing module for live data display.
• Designed Arduino and Raspberry Pi based circuits, integrating devices like TFT touch display, 802.11 module, ADCs and analog switch matrix microchips by employing bus communication protocols like SPI and I2C.
Honors & Awards
-
CrushIT Award Winner of 2020
IBM
-
Master Inventor
IBM
-
Manager's Choice Award - 2019 1H
IBM
-
Tenth Plateau Invention Achievement Award
IBM
-
Manager's Choice Award - 2018 1H
IBM
-
Invention Development Team Award
IBM
-
Second Plateau Invention Achievement Award
IBM
-
First Plateau Invention Achievement Award
IBM
-
Patent Application Invention Achievement Award
IBM
-
Winner of Best Social Good Hack - mark43
DevFest 2016 Hackathon
-
Merit Scholarship - 3 years
Netaji Subhas Institute of Technology, Dwarka, Delhi
-
Finalist - Honeywell Young Innovators Challenge
https://sites.google.com/site/younginnovatorchallenge/results
-
Computer Internet Security Expert
Innobuzz Knowledge Solutions
Test Scores
-
MS in Electrical Engineering
Score: GPA : 3.38
-
Bachelor's in Engineering
Score: 72.49% (GPA: 3.68/4)
-
AISSCE(Class XII) CBSE
Score: 89.8%
-
AISCE (Class X) CBSE
Score: 86.2%
Languages
-
English
Native or bilingual proficiency
-
Hindi
Native or bilingual proficiency
-
Punjabi
Professional working proficiency
Organizations
-
NCTEC
Member / Program committee Chair
-
NSIT Rotoract
Active Student member
-
IEEE
Student Member
Recommendations received
22 people have recommended Shikhar Join now to view
More activity by Shikhar
An update, 3 years into this next chapter of my career: 3 years ago in May 2021 — after nearly 20 years in traditional tech roles as an engineer and…
Liked by Shikhar Kwatra
xLSTM As LLM: Extended Long Short-Term Memory 💽 This paper introduces an enhanced version of the LSTM architecture that aims to improve its ability…
Shared by Shikhar Kwatra
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More