International Business Machines Corporation

Kyndryl, Inc.

Swaminathan Balasubramanian

Cheranellore Vasudevan

Radha Mohan De

Brian M O'Connell

Indrajit Poddar

Ashley Dallas Delport

Keith R Walker

Rick A Hamilton, Ii

Ravi Prakash Bansal

Andrew R Jones

Pierre C Berlandier

Gregory J Boss

Rajiv V Joshi

Mark Bell

Sibasis Das

John J Thomas

Santosh S Borse

Avijit Chatterjee

Jason Richard Malinowski

Thomas G Lawless, Iii

Sarbajit Kumar Rakshit

Peeyush Jaiswal

John Hans Handy-Bosma

Priyansh Jaiswal

Agueda Martinez Hernandez Magro

Kenneth Wayne Roberson

Mamnoon Jamil

Radha Radha Mohan De

Saswati Maji

Sandip Dnyaneshwar Mahajan

Derek P Bagwell

Robert A Morgan

Jian Wu

Krisha Banerjee

Clifford A Pickover

Neal Martin Keller

Jeffrey David Amsterdam

Brian James Snitzer

Bruce David D'Amora

Matthew Mitchell Lobbes

Renganathan Sundararaman

Karina Elayne Kervin

Genevieve Van Den Boer

Rajib Bhattacharya

Haaris A Shaikh

Jay K Strosnider

Detecting duplicate tables by converting relational tables to knowledge graphs; and mapping nodes in the knowledge graphs to sources in the relational tables. The method may further include applying graph matching to the knowledge graphs; and assessing degree of matching between matched knowledge graphs.

Detecting duplicate tables in data lake databases

A method, computer system, and a computer program product for historical image search is provided. The present invention may include receiving an image search criteria. The present invention may also include generating a composite image search query based on the received image search criteria. The present invention may further include searching for a matching historical image based on the generated composite image search query. The present invention may also include returning a historical image search result.

Historical image search

Mechanisms are provided for optimizing remuneration for computing services. Computing services are registered which stores registration data comprising remuneration associations between computing services and consumers. A hierarchical computer model is generated based on the registration data, where the model represents dependencies between consumers and providers of computing services. For a service request from a consumer, each transaction with each computing service in a service invocation chain associated with the service request is identified and a cost of each transaction is calculated. The calculated cost of the service invocation chain is optimized based on applying an optimization algorithm to a cost function applied to a selected portion of the hierarchical computer model corresponding to the service invocation chain.

Computer service invocation chain monitoring and remuneration optimization

A method for a story fill authentication process includes, responsive to receiving a first authentication request to authenticate a user, displaying a first generated story with one or more obfuscated portions, where the first generated story is based on event data associated with a first previously captured event and additional data utilized to enrich the event data for the first previously captured event. The method also includes, responsive to determining text provided for the one or more obfuscated portions of the first generated story at least meets a comparison threshold level to a first complete generated story based on a semantic comparison, granting the user access to a resource associated with the first authentication request.

Generated story based authentication utilizing event data

A client device can receive, from a data processing system, first data, the first data selected by the data processing system based on second data generated by at least one IoT device with which the client device is associated, and the first data indicating at least one advertisement. The client device can access the at least one advertisement using the first data indicating the at least one advertisement. The client device can add the at least one advertisement to the content accessed by the client device. The client device can present the content with the at least one advertisement added to the content.

Personalization of advertisement selection using data generated by IoT devices

In an approach to transient identification generation, one or more computer processors receive a request for one or more identification artifacts associated with a user from a requesting entity. One or more computer processors determine a first set of identification artifacts. One or more computer processors transmit the first set of identification artifacts to the requesting entity. One or more computer processors determine whether the first set of identification artifacts is an approved response to the request for the one or more identification artifacts. In response to determining the first set of identification artifacts is the approved response to the request for the one or more identification artifacts, one or more computer processors generate a transient identification that includes the first set of identification artifacts. One or more computer processors transmit the transient identification to the requesting entity.

Transient identification generation

A method according to one embodiment is for transmitting a message. In one embodiment, one or more processor of a messaging system identifies a certain message of the messaging system, the certain message being a blocked message intended for transmission to a certain user of the messaging system. In one embodiment, the one or more processor further determines one or more of a popularity of content of the certain message and a relevance of content of the certain message to the certain user, the certain user being a prospective recipient of a message. In one embodiment, the one or more processor transmits a message to the certain user responsively to a result of the determining of the one or more of a popularity of content of the certain message and a relevance of content of the certain message to the certain user of the messaging system.

Transmittal of blocked message notification

Embodiments for scheduling the delivery of multiple medications to a patient by a processor. Any interactions between the multiple medications or between one of the multiple medications and a type of food are considered to construct an analytical model to identify an optimized schedule for delivery of the multiple medications to the patient. The optimized schedule is adjusted in real time as additional information concerning the multiple medications and the type of food are introduced, and delivered to a dispensing device associated with the one of the multiple medications.

Optimization of medicines delivery

A method for event data fencing includes initializing a media monitoring algorithm, wherein the media monitoring algorithm scans for one or more words relating to a data breach in publicly available reports. Responsive to identifying a report from the publicly available reports relating to a vulnerability associated with the data breach, the method determines whether the vulnerability is associated with event data utilized for an authentication process. Responsive to determining the vulnerability is associated with a portion of the event data utilized for the authentication process, the method fences the portion of event data associated with the vulnerability, wherein fencing prevents the portion of event data from being utilized in the authentication process.

Event data fencing based on vulnerability detection

A tool for inter-vehicle communication. The tool determines whether contextual data and journey information for a user justifies a notification. Responsive to a determination that the contextual data and the journey information justifies the notification, the tool generates the notification. The tool generates a trust score for the notification. The tool determines one or more vehicles located within a pre-defined proximity of the user. The tool transmits the notification and the trust score to the one or more vehicles located within the pre-defined proximity of the user.

Inter-vehicle communication

A method for a two-factor authentication process includes, responsive to determining a first user authentication was prompted in a first application on a first device associated with a user, identifying a second application on a second device based on a user profile associated with the user. The method identifies a first event from a plurality of events that previously occurred in the second application in a select time frame, wherein the first event relates to a first action performed by the user in the second application. The method generates an authentication question based on the first event, wherein the authentication question is a second user authentication. Responsive to determining an answer provided by the user to the authentication question is correct, the method grants access to the first user authentication prompt.

Multi-factor authentication utilizing event data

Method and apparatus for generating profiles using machine learning and influencing online interactions are provided. The methods include generating a user profile specifying a plurality of attribute values for a plurality of principle attributes, by processing a corpus of electronic documents using a first trained machine learning model. In an embodiment, the method further comprises generating a provider profile specifying a plurality of attribute values for the plurality of principle attributes, for each of a plurality of providers, by processing a respective corpus of electronic documents associated with each respective provider using a second trained machine learning model. A plurality of match coefficients based on comparing the user profile and the plurality of provider profiles are determined. Finally, one or more online interactions between the user and the target provider are influenced based on the determined match coefficients.

Building and matching electronic standards profiles using machine learning

Peer-to-peer transfer of compute function state in an edge computing ecosystem is provided. An availability score corresponding to a mobile computing device is received. It is determined whether the availability score is less than an availability score threshold. In response to determining that the availability score is less than the availability score threshold, departure coordinates of the mobile computing device from the edge computing ecosystem and departure time are determined. At least one peer mobile computing device is identified having a corresponding availability score greater than the availability score threshold, a corresponding security profile that at least matches a security profile of the mobile computing device, and a corresponding compute power capability that at least matches a compute power capability of the mobile computing device. A compute function state of the mobile computing device is transferred to the at least one peer mobile computing device.

Peer-to-peer transfer of edge computing based on availability scores

Method and apparatus for generating profiles using machine learning and influencing online interactions are provided. The methods include receiving, from a first user of a plurality of users, a first set of electronic documents, where each electronic document in the first set of electronic documents corresponds to a respective user in the plurality of users. The methods also include identifying a plurality of user profiles, where each of the plurality of user profiles was generated by processing a corpus of electronic documents associated with each respective user using a first trained machine learning model. The methods include determining a plurality of match coefficients, based on comparing a plurality of user profiles associated with each respective user in the plurality of users, filtering the first set of electronic documents based on the plurality of match coefficients, and providing the filtered first set of electronic documents to the first user.

Building and matching electronic user profiles using machine learning

Provided is a method for disambiguating an audio component extracted from audiovisual content. Audiovisual content is identified. The audiovisual content includes an audio component and a video component. An ambiguous expression is detected in the audio component. An object referenced by the ambiguous expression is identified in the video component. A verbal description of the object is generated. The verbal description is injected into the audio component to generate a modified audio component.

Disambiguation of audio content using visual context

Growing community of inventors

Troy, MI, United States of America

Swaminathan Balasubramanian