International Business Machines Corporation

Sushain Pandit

Martin Oberhofer

Ivan Matthew Milman

Shikhar Kwatra

Charles Daniel Wolfson

Brian Patrick Byrne

Dan Jeffrey Mandelstein

Sarbajit Kumar Rakshit

Su Liu

Brent P Gross

Fang Wang

David Alec Selby

Paul N Krystek

Komminist S Weldemariam

Vijay Ekambaram

Jeremy R Fox

Vittorio R Castelli

Cheng Xu

Abdigani Mohamed Diriye

Calvin Stacy Powers

Otis Smart

Harikumar Venkatesan

Liam S Harpur

Kuntal Dey

Florian Pinel

Smitkumar Narotambhai Marvaniya

Wayne Francis Tackabury

Qin Qiong Zhang

Subhajit Bhuiya

Guenter Anton Sauter

Jitendra Singh

Donald Andrew Padilla

Michael R Self

Subramanian Palaniappan

Frederick L White

Robert Kern

Dmitriy Fot

Charles K Shank

Yogendra K Srivastava

Kalaivanan Saravanan

Joerg Rehr

Clifford A Pickover

Corville O Allen

Jonathan D Dunne

Zachary A Silverstein

Peeyush Jaiswal

Romelia H Flores

Raghuveer Prasad Nagar

Vinod A Valecha

Tiberiu (Jim) Suto

Priyansh Jaiswal

Mohit Jain

Jagabondhu Hazra

Indervir Singh Banipal

Andrew T Penrose

Fenglian Xu

Adam Reudeau Holley

Lloyd Alan Treinish

Richard Daniel Gunjal

Saurabh Yadav

Nishtha Madaan

Mukul Tewari

Joseph True

Geeth Ranmal De Mel

Andrew Kinai

Saurav Basu

Lan Ngoc Hoang

Daniel C Wolfson

Corey A Stubbs

Krishna Teja Rekapalli

Dean Daskalantonakis

Wesley Richards

Steven Lockwood

Kovit Nisar

Rory O Donnell

A method, a structure, and a computer system for playback optimization. The exemplary embodiments may include transcribing an audio stream at normal and at high playback speed, as well as determining whether portions of the audio stream are unintelligible at the high playback speed. The exemplary embodiments may further include reducing a speed of the high playback speed for portions determined unintelligible, and transcribing the audio stream with the unintelligible portions played at the reduced high playback speed. The exemplary embodiments may further include determining whether the unintelligible portions are still unintelligible at the reduced high playback speed and, if not, identifying the reduced high playback speed as an optimal playback speed of the unintelligible portions. The exemplary embodiments may additionally include, in response to receiving a request to play the audio stream at the high playback speed, playing the unintelligible portions at the optimal playback speed.

Audio playback and captioning

In a method for intelligently executing predictive simulator, a processor may input a previous input vector of conditions for a predictive simulator collected at a first time into a machine-learning (ML) model. A processor may input a current input vector of conditions for the predictive simulator collected at a second time into the ML model. A processor may determine using the ML model, a binary similarity index. The binary similarity index represents a prediction of similarity between a first output from the predictive simulator based on the previous input and a second output from the predictive simulator based on the current input.

Intelligent execution of compute intensive numerical simulation models

In an approach for enabling a user to visualize a transcript of a discussion via a head-mounted AR device and to selectively copy one or more parts of the transcript that is contextually relevant for inclusion in a new document file or a previously created document file, a processor captures audio of a spoken content of a first participant of a discussion via an AR device worn by a user. A processor analyzes the audio of the spoken content of the first participant. A processor converts the audio of the spoken content of the first participant to text to create a transcript. A processor creates a visualization of the transcript. A processor presents the visualization of the transcript to the user via the AR device. A processor enables the user to copy one or more parts of the transcript into a document file via a selection support.

Selective inclusion of speech content in documents

In an approach to meeting summarization for disconnected participants, one or more computer processors detect a start of a collaborative meeting. One or more computer processors monitor a connection to the collaborative meeting of each of one or more participants. One or more computer processors determine a first participant of the one or more participants has disconnected from the collaborative meeting. One or more computer processors record the collaborative meeting. One or more computer processors determine the first participant has re-connected with the collaborative meeting. One or more computer processors synthesize, a first summary of a portion of the collaborative meeting missed by the first participant. One or more computer processors to synthesize, a second summary of a portion of the collaborative meeting recorded after the first participant re-connected with the collaborative meeting. One or more computer processors present the first summary and the second summary to the first participant.

Meeting content summarization for disconnected participants

A method includes: receiving, by a processor set, a cloud architecture document; converting, by the processor set, the cloud architecture document to a graph including nodes; determining, by the processor set, cloud environment information for the nodes by analyzing textual content of the nodes using natural language processing; determining, by the processor set, additional cloud environment information for the nodes using domain ontology data; and creating, by the processor set, an Infrastructure as Code (IaC) document based on the cloud environment information and the additional cloud environment information.

Converting an architecture document to infrastructure as code

A method, system, and computer program product for identifying reference data values in a source data set. The method may include inputting a block of attribute values to a predefined machine learning model. The method may also include receiving an indication of a presentation layout of the block of the attribute values and an associated reference data extraction method. The method may also include determining a reading direction of the block of values. The method may also include identifying one or more inspection areas in the reading direction of the block of values. The method may also include determining sets of the one or more inspection areas that share a common presentation feature. The method may also include identifying tokens in an inspection area. The method may also include determining if the inspection area includes reference data values. The method may also include outputting the reference data values.

Identifying reference data in a source data set

In an approach for routing a user analytic request to a registry of analytics models and data sources and operationalizing the user analytic request, a processor receives a user analytic request. A processor applies natural language processing to the user analytic request. A processor associates a first set of vectors and a second set of vectors from the user analytic request to one or more analytics models and data sources by utilizing a plurality of domain-specific ontologies and a plurality of knowledge bases. A processor performs a semantic search for one or more concepts in the one or more analytics models and data sources. A processor receives the one or more concepts found in the one or more analytics models and data sources. A processor selects an analytics model to process the user analytic request. A processor executes the analytics model. A processor outputs a result to the user.

Context-driven analytics selection, routing, and management

Communications are managed and prioritized by a machine learning process in a vehicle with automated driver assistance. It is detected that the vehicle is currently being operated in manual mode by a human driver. It is detected that a telecommunication device located within the vehicle is receiving a communication. The communication is classified according to a priority. The communication is acted upon based on the priority classification. The driver state is assessed at the conclusion of the communication. The vehicle returns to the manual driving mode if the driver state is compatible with the manual driving mode.

Communication prioritization in vehicles with automated driver assistance

A computer-based method, system, and computer program product for automatically identifying significant events for food traceability. The method may comprise receiving a series of events from an agriculture supply chain entity, automatically determining, at a machine learning model of an event analysis module, one or more events in the series having a significance for food traceability greater than a threshold, and automatically reporting the one or more events to a ledger.

Determining significant events within an agribusiness system

Embodiments relate to a system, program product, and method for leveraging cognitive systems to facilitate the bilateral contextual alignment of textual material and associated diagrammatic matter. The system, computer program product, and method disclosed herein facilitate leveraging a trained cognitive system to understand textual and diagrammatic patterns established by a user, and to learn to draft and edit such textual material and associated diagrammatic matter based on user behavior and preferences. The cognitive system is trained to bilaterally translate between textual material and diagrammatic matter, where the cognitive system includes natural language processing (NLP) features, diagram analytics features, and one or more user profiles. Machine-drafting of a document is achieved through the trained cognitive system such that first textual material and first diagrammatic matter, including first pictorial objects and first diagrammatic textual matter, are bilaterally translated into second textual material and second diagrammatic matter, which are combined and contextually aligned.

Contextual diagram-text alignment through machine learning

A method, a computer program product, and a computer system determine a customized augmented reality workout based on a cognitive state of a user. The method includes determining a current cognitive score of the user based on physical descriptions. The method includes determining the cognitive state of the user based on the current cognitive score. The cognitive state is indicative of a stress level being experienced by the user. The method includes determining the customized augmented reality workout from a plurality of augmented reality workouts based on the cognitive state of the user. The method includes generating the customized augmented reality workout to include at least one virtual component and incorporating at least one feature of real world surroundings. The method includes presenting the customized augmented reality workout to the user via an augmented reality device worn by the user.

Cognitive based augmented reality workout

A method for protecting content of online conversational content. The method provides for scanning content of an online conversational exchange between a first device and a second device. A sensitive object included in the content of the online conversation exchange is identified, based on object type information accessible from a protection policy included in respective user profiles. A pseudonymized-object-holder is assigned to the identified sensitive object according to the protection policy of the respective user profiles, and the identified sensitive object identified in the content of the online conversation exchange and stored on both the first device and the second device is replaced with a pseudonymized-object-holder, based on the sensitive-object protection policy of the respective user profiles.

Pseudonymizing sensitive objects in online content exchanges

A method, computer system, and a computer program product for problem solving using selected datasets from an internet-of-things system are provided. A trigger device in an internet-of-things system may be designated. The internet-of-things system may include the trigger device and internet-of-things devices. A trigger message may be received from the trigger device. A problem may be determined based on the trigger message. Data transmissions may be requested from the internet-of-things devices. Data may be received via the data transmission. The data transmission may be ended. A problem contributor may be identified by inputting the received data and the determined problem into a machine learning model. A problem response for responding to the problem may be generated. The generating may include inputting the problem contributor to the machine learning model. The problem response may be performed.

Problem solving using selected datasets of internet-of-things system

Cognitive pre-loading of referenced content in electronic mail (e-mail) messages includes determining, using computer hardware, metadata for an electronic message directed to a user, detecting, using the computer hardware, a universal resource locator (URL) within a body portion of the electronic message, determining, using the computer hardware, a content type for the URL, and performing, using the computer hardware, natural language processing on the electronic message to determine an action importance corresponding to the URL. At least a portion of content specified by the URL can be pre-fetched from a data processing system based on the metadata, the content type of the URL, and the action importance. The electronic message and at least a portion of the content can be provided to a client device of the user.

Cognitive pre-loading of referenced content in electronic messages

Embodiments of the present invention describe creating a mesh network of autonomous vehicles. Embodiments describe, for a set of autonomous vehicles configured to perform joint tasks, creating a plurality of subset networks of autonomous vehicles based on geographical proximity, wherein the plurality of subset networks of autonomous vehicles overlap to form a mesh network of autonomous vehicles, and wherein the mesh network of autonomous vehicles and the subset networks of autonomous vehicles support a non-centralized task assignment feature to assign tasks to autonomous vehicles in the mesh network. Additionally, embodiments describe sending a status signal to a first autonomous vehicle in the mesh network via a second autonomous vehicle in the mesh network, wherein the first autonomous vehicle is in a same subset network of autonomous vehicles as the second autonomous vehicle.

Dynamic autonomous vehicle mutuality support for autonomous vehicle networks

Growing community of inventors

Austin, TX, United States of America

Sushain Pandit