International Business Machines Corporation

Laurence J Plant

Stefan Harrer

Christopher J Pavlovski

Michael Sean Brown

Sean Rory Costello

James David Cleaver

Josef Reisinger

Canio Cillis

Jochen Kappel

Julian De Hoog

Michael S Brown

Tania Ezra

Christopher James Poynter

Leon Gouletsas

A method may include obtaining a first data set of a first user present at a venue. The method may also include obtaining venue characteristic data corresponding to a set of characteristics of the venue. The method may also include obtaining first venue condition data corresponding to a first set of venue conditions. The method may further include generating a first evacuation route based at least in part on the first data set. The method may also include generating a first timing instruction corresponding to the first evacuation route and based at least in part on the first venue condition data. The method may further include providing the first evacuation route and the first timing instruction to the first user.

Personalized venue evacuation plan

The present disclosure teaches a solution for a user customizable abstraction layer for tailoring all operating system, application, and web based interfaces. The interface differs from conventional user interfaces by presenting a dynamic interface which can enable user access across all domains and applications with which the user can interact. The interface can be dynamically built as a user interacts with clients (e.g., devices/applications). Clients can utilize common usage patterns, installed application, installed themes, personal information, and the like, to create a highly customized adaptive user designed and modifiable interface.

Dioramic user interface having a user customized experience

A method, system, and computer program product for enabling a digital search is provided. The method includes monitoring, in response to receiving authorization from a user, Web search results of a user, social media accounts, and media sources. In response data indicating an importance of world related factors with respect to the user is received. The data is analyzed with respect to results of the monitoring and an associated metadata model and modeling software application personalized for the user is generated. A search request associated with initiating a Web search for a product or service is initiated and the modeling software application is executed with respect to the metadata model and Web search. Personalized Web search results are generated and presented to the user. The personalized Web search results indicate entities associated with providing the product or service for the user via Internet based commerce Websites.

Digital search results generation

Methods, computer program products, and systems are presented. The methods include, for instance: obtaining a style feed including a plurality of original works by an author. An author-style model for the author is built based on the style feed by use of a selected neural network, and a publication is generated in the style of the author based on the author-style model.

Cognitive natural language generation with style model

A method and system for improving a machine learning technical support applicant selection process is provided. The method includes retrieving and analyzing, in real time, job role data, resume based data, and sensor data associated with a plurality of individuals. A specialized memory repository is generated within a specified portion of a hardware memory device and results of the analysis are stored within the specialized memory repository. Self-learning software code for executing a job applicant selection process is generated and a plurality of current resumes and previously generated time-based resumes are retrieved for current job applicants for a specified position. The current resumes are analyzed with respect to the previously generated time-based resumes and associated inconsistencies are determined. A first job applicant is selected and the self-learning software code is modified.

Machine learning technical support selection

A cognitive coaching method; detecting one or more real-time specific action occurrences during one or more game phases; generating, with a heat map generator, one or more real-time heat maps with one or more specific action label based on the detected one or more real-time specific action occurrences; training a heat map image classifier with an analytics module; transposing one or more historical heat maps comprised in a historical heat map library onto the one or more generated real-time heat maps, in real time, to generate strategic insight on one or more opposing teams; analyzing, using a predictive module, a future potential outcome based on the one or more transposed heat maps; and signaling, to a user, a recommended corrective action deployment strategy based on the analysis of the future potential outcome.

In-game strategic insight with real-time heat map generation and historical heat maps

An approach is provided for optimizing a deployment pattern. Precedent system(s) that have components that match components in an initial deployment pattern of a new system are identified. A historical performance of the precedent system(s) is determined to not satisfy target non-functional requirements (NFRs) and target service levels of the new system. Responsive to determining that the historical performance did not satisfy the NFRs and the target service levels, the components in the initial deployment pattern are modified and a new deployment pattern for the new system is generated so that (i) the new deployment pattern includes the modified components and (ii) a performance of the new system using the new deployment pattern is likely to satisfy the target NFRs and the target service levels. A recommendation for deploying the new system using the new deployment pattern is generated.

Constructing and enhancing a deployment pattern

Methods, computer program products, and systems are presented. The methods include, for instance: obtaining sample essays, sample annotations corresponding to the sample essays, and a subject content for building a subject domain comprehension model and an essay annotation model, by use of one or more neural network. The nodes of the subject domain comprehension model and the essay annotation model are interconnected based on respective relevancies for automatically annotating student works according to a standard of review corresponding to submitting students.

Cognitive essay annotation

A method and system for predicting system states is provided. The method includes receiving a first reference model associated with a first operational attribute of a system from a first integrated circuit internally comprising a first processing circuit and a first sensor measuring a first parameter of the system. Additionally, a second reference model associated with a second operational attribute of the system is received from a second integrated circuit internally comprising a second processing circuit and a second sensor measuring a second parameter of the system. A combination reference model based on the first reference model and the second reference model is generated and a predicted future state and associated operational attributes for the system are determined based on the combination reference model.

Sensor based system state prediction

Remote access of a service provider to a secure enterprise computing environment through a firewall through the use of tokens.

Secure access to an enterprise computing environment

Hybrid device for viewing media content: a media player; a non-transitory storage medium; and a processer to perform functions: exception detection and handling to monitor a rate of media streaming and to identify when viewing of the media content is or predicted to be interrupted; a peer interaction manager to broadcast a request to peers for media segments of the media content when viewing of the media content is or predicted to be interrupted; a security manager to send a request to a content distribution network to start streaming of the media content, to receive a session key from the content distribution network, and to present the session key to the peers; a peer media manager to receive the media segments from the peers and cache the media segments; and a media segment manager to stitch the media segments from different peers into a single continuous media file.

Media streaming using hybrid P2P and client-server distribution of content

A method and system for improving a digital media content transfer blocking process is provided. The method includes detecting a request for access to digital media content from a digital media provider system. Network traffic across the network is identified and it is verified that the digital media content is comprised by the network traffic. In response, investigation platform software code is executed and a database system associated with the digital media provider system is queried. A system determines if a legitimate connection based relationship exists between a user device and the digital media provider system and the digital media content is monitored. The digital media content is determined to be unauthorized for use by the user and in response the digital media content is disabled from executing functions with respect to the user device of a user.

Digital media content distribution blocking

A physical network address of one of a set of wireless customer-premise monitoring devices is translated to a shared logical device physical network address in response to receipt from an application server of a first data packet addressed to the physical network address. The shared logical device physical network address forms a logical device group and physically addresses each of the set of wireless customer-premise monitoring devices over a single wireless communication channel. The physical network address of the one of the set of wireless customer-premise monitoring devices is embedded within the first data packet addressed to the shared logical device physical network address. The first data packet is sent to the shared logical device physical network address.

Multi-device monitoring and control using intelligent device channel sharing

A first message resilience client device receives a message and a request to deliver the message, on behalf of a remote client/server-based client application that originated the message, to a client/server-based server application that provides outgoing client/server messaging services to the remote client/server-based client application that originated the message. In response to determining that the connection to the server device that executes the client/server-based server application is not currently possible with any available connection, the message is stored locally for one of later delivery to the client/server-based server application and propagation of the message to another message resilience client device on behalf of the remote client/server-based client application.

Resilient messaging infrastructure

An approach is provided for optimizing a deployment pattern. Event and configuration data of precedent systems is aggregated. Based on the aggregated data, a metadata model of the precedent systems is generated. Components of an initial deployment pattern of a new system are identified. Target non-functional requirements (NFRs) and target service levels of the new system are determined. Based on the metadata model, precedent system(s) that have components that match or are similar to the components in the initial deployment pattern are identified. A performance of the precedent system(s) is compared to the NFRs and target service levels. A new deployment pattern for the new system is generated. The new deployment pattern includes a modification of the components identified in the initial deployment pattern. A recommendation for deploying the new system using the new deployment pattern is generated.

Growing community of inventors

North Baldwyn, Australia

Laurence J Plant