International Business Machines Corporation

Kyndryl, Inc.

Nuance Communications, Inc.

Amit Anil Nanavati

Nitendra Rajput

Ashish Kundu

Arun Kumar

Anupam Jain

Dipanjan Chakraborty

Sheetal K Agarwal

Sougata Mukherjea

Natwar Modani

Kuntal Dey

Seema Nagar

Biplav Srivastava

Danny Soroker

Raghavendra Singh

Nilanjan Banerjee

Koustuv Dasgupta

Clifford A Pickover

Jenny S Li

Komminist S Weldemariam

Birgit Monika Pfitzmann

Ramasuri Narayanam

Ravi Kothari

Maryam Ashoori

Palanivel Andiappan Kodeswaran

Balaji Viswanathan

Jitendra Ajmera

Shalini Kapoor

Om Dadaji Deshmukh

Rahul Singh

Elezaveta Koumpan

Anil K Maddipatla

Ruta Mehta

Subramanya Kabbinahithlu Narayana Prasad

Erik Rueger

Manish Anand Bhide

Manoj N Kumar

Sachindra Joshi

Srikanth Govindaraj Tamilselvam

Pankaj Satyanarayan Dayama

L Venkata Subramaniam

Archan Misra

Raghuram Krishnapuram

Gyana Ranjan Parija

Manish Kataria

Steven A Keller

Amrita Saha

Priyanka Agrawal

Pralhad Dinesh Deshpande

Sayandeep Sen

Apurva Kumar

Vishal Singh Batra

Amir Ronen

Parul Alok Mittal

Amit Mohan Mangalvedkar

Pankaj Kankar

Krishna Prasad Chitrapura

Rajan Kumar

Vinod Vinayak Mankar

Nicolò Sgobba

Manish P Kurhekar

Guillermo Rodriguez De Vera Beltri

Chourasia Abhishek Kumar

Karunakaran Gajulu Narasimhalu

Megha Gupta

Mayur Saxena

Natwar Modan

Jason Hardy

Usha N Raghavan

Narendra Kumar Agrawal

Zheng Zhu

Mohit Kumar

Chinmoy Dutta

Suraj Kumar Jaiswal

Pasumarti V Kamesam

A computer-implemented method includes: receiving, by a computer device, an electronic document having labels; predicting, by the computer device, a user will reject the labels; determining, by the computer device and in response to the determining the user will reject the labels, that a subset of labels of the labels violate association rules; marking, by the computer device, the subset of labels which violate the association rules for validation; prioritizing, by the computer device, the subset of labels which violate the association rules; and rendering, by the computer device, the subset of labels which violate the association rules in view of priority.

Interactive validation of annotated documents

The present disclosure relates to a method of coordinating personal assistants. The method comprises: sending a request to a set of personal assistants based on a received query. Assistant responses may be received in response to the request from respective personal assistants of the set of personal assistants. Each assistant response of the assistant responses comprises a confidence score of the respective assistant response. The confidence scores may be adjusted to a common scale. The adjusted confidence scores may be weighted using domain-specific weights of the personal assistants in accordance with the domain of the query. One or more assistant responses of the received assistant responses may be selected based on the weighted scores and the selected assistant responses may be provided to the user.

Virtual assistants harmonization

A method and associated systems provide real-time response to a request received from a real-time system like a self-driving vehicle or a device that communicates interactively with its environment. The response is selected from a set of candidate feasible responses by a group of computerized agents that each sort the feasible responses in order of that agent's specific preferences, based on that agent's particular priorities or expertise. The agents then reconcile their differences through an iterative procedure. During each iteration, each agent decides whether to retain its current preferences or to adopt the preferences of another agent. This decision is made by determining which preferences are most similar to that agent's own initial preferences, and by which preferences would be most useful in helping to achieve that agent's particular goals. When the agents reach consensus, the group's most-preferred response is returned quickly enough to provide real-time, interactive response.

Real-time multi-agent response based on a preference-based consensus

A method, apparatus, and non-transitory computer readable medium for performing joins on data from hierarchical databases are described. The method, apparatus, and non-transitory computer readable medium may provide for receiving one or more search results from each of a plurality of hierarchical databases, identifying one or more matching fields from each of the search results, joining the search results according to a set of rules for processing related fields to the matching fields, wherein the related fields comprise sibling fields, neighbor fields, ancestor fields, descendant fields, or any combination thereof, and generating one or more combined search results based on the joining.

Supporting joins in directed acyclic graph knowledge bases

A parameter of a first checker in a profile of the first checker is configured. The parameter comprises a restriction on contents of a modification to a dataset. The modification comprises a write to a portion of the dataset. The dataset is modifiable by a plurality of authors. The first checker has a first checker owner. A first modification to the dataset is validated against a restriction enforced by the first checker according to the parameter. The first modification has a modification owner. Responsive to the validating detecting that the first modification is acceptable, the first modification is allowed.

Consistency enforcement in multi-author documents

Methods and arrangements for data-driven discovery of collaborative models in an internet of things. At a device in the internet of things setting, data are collected on ambient conditions, and the collected data are stored at the device. At least one other device in the internet of things is communicated with to create and record collaborative states relative to the device and the at least one other device. Other variants and embodiments are broadly contemplated herein.

Discovering capabilities of entities in an internet of things setting

Mechanisms are provided for implementing a conversation monitoring system. The conversation monitoring system monitors a conversation between at least two participants and extracting, by the conversation monitoring system, key terms present in communications between the at least two participants. The conversation monitoring system generates a provenance graph data structure based on the extraction of the key terms where the provenance graph data structure has speaker nodes representing the at least two participants, term nodes representing the key terms present in the communications, and edges connecting nodes in the provenance graph. The conversation monitoring system analyzes the provenance graph data structure to identify a relative ranking of the key terms within the conversation and generates an output representing content of the conversation based on the relative ranking of the key terms.

Cognitive ranking of terms used during a conversation

Methods and arrangements for determining one or more actions needed to execute a task communicated to a first device in an Internet of Things. At a first device in the Internet of Things, there is received a communication associated with a task. A data store is consulted for data associated with the task, and thereupon there is determined the present capability of the first device for executing the task collaboratively with at least one other device in the Internet of Things. The at least one other device is communicated with, to assess capability for executing the task collaboratively with the at least one other device. There are thereupon determined one or more actions of the first device that need to be taken to execute the task. Other variants and embodiments are broadly contemplated herein.

Actuation using collaboration models in an internet of things setting

One embodiment provides a method, including: creating, for a user, a sound desirability index comprising (i) a plurality of sounds and (ii) desirability of the sound to the user with respect to each of the plurality of sounds, wherein the sound desirability index is created in response to the user hearing a sound and the environment of the user when hearing the sound; receiving an audible input within hearing proximity of the user; identifying the current environment of the user; determining the desirability of the audible input to the user by accessing the sound desirability index, and determining the desirability of the audible input based upon a sound in the audible input and the current environment of the user; and modifying a characteristic of at least a portion of the audible input based upon the desirability of the audible input to the user.

Filtering sound based on desirability

Provided is a computer-implemented method for automatically preserving privacy in a user interface. The method includes detecting privacy objects that are being presented via a UI of a computing device to a first user. A security status of the UI is evaluated based at least in part on analyzing audio captured by an audio capture device of the computing device. A privacy mode of the UI that executes privacy actions that are associated with the detected privacy objects is activated based on determining that the security status is in an unsecure state. The privacy mode is deactivated based on determining that the security status has changed from the unsecure state to a secure state, in which determining that the security status has changed from unsecure to secure is based on analyzing the audio captured by the audio capture device of the computing device.

Cognitive widgets and UI components for preserving privacy and security

A first electronic message being widely disseminated can be identified. First content of the first electronic message can be analyzed and a determination can be made as to whether the first content satisfies at least a first criterion. Responsive to determining that the first content satisfies at least the first criterion, a second electronic message including second content can be automatically generated. Communication of the second electronic message to a first set of people can be initiated, and responses by the first set of people to the second electronic message can be monitored. A determination can be made as to whether the responses satisfy at least a second criterion. Responsive to determining that the responses satisfy at least the second criterion, the second electronic message can be communicated to at least a second set of people.

Responding to an electronic message communicated to a large audience

An embodiment may be a cognitive event based file sharing system. The embodiment may include receiving a file share request by a user, wherein the file share request comprises a file selected for access by intended file recipients. The embodiment may include determining a storage capacity of a computing device of the user. The embodiment may include determining a storage capacity of computing devices of the intended file recipients. The embodiment may include allocating the file based on the storage capacity of the computing device of the user, the storage capacity of the computing devices of the intended file recipients, and a file management policy. The embodiment may include providing access to the file to the intended file recipients. The embodiment may include removing a portion of the file from at least one of the computing devices of the intended file recipients based on the occurrence of a specified event.

Cognitive event based file sharing system for social software

Embodiments of the invention provide a non-web based social networking system enabling users having limited Internet connectivity and/or limited access to high end mobile computing devices a cross networking opportunity with web-based social networking sites. Embodiments of the invention provide for additional communication modalities between social networking users, including but not limited to non-web based communication between the social networking users.

Systems and methods for networking across web based and telecommunication based portals

Growing community of inventors

New Delhi, India

Amit Anil Nanavati