International Business Machines Corporation

Chanchal Saha

Richard Burton Finch

Jayapreetha Natesan

Kanayo Okonji

Michael Lapointe

Gerald Stanquist

Jason Hirst

Oswald J Mantilla

Warren Boldrin

Jason McAllister

Rajaram B Krishnamurthy

Natesan Venkateswaran

Marc Henri Coq

William J Green

Thembani Togwe

Thomas Ward

Brad Hoover

Jeffrey Willoughby

Rogelio Fernando Gutierrez Valdes

Christian Nugraha

Anthony George Gorey

Neeraj Madan

Oscar Alejandro De La Torre

Sumant Singh Rana

Gersain Mesa

Teik Seng Poh

Richard Laviano

James Hoey

Steven Holbrooks

Robert Edge, Jr

Mark C Booth

Restricting use of a selected input in recovery from system failures. A testing system obtains, based on failure of a system, an indication of one or more tasks to be performed. The testing system automatically determines whether a user is to perform one or more actions to alter execution of the one or more tasks. The automatically determining is based on an indication of whether a knowledge transfer score of the user meets at least one predefined criterion. Based on knowledge transfer score meeting the at least one predefined criterion, the one or more actions are performed.

Restricting use of selected input in recovery from system failures

A cognitive test advisor facility is provided to identify repair action recommendations. The facility receives a user input regarding a test failure, and generates an expanded search input by identifying additional, alternative words for the user input. A database(s) is searched for test failure records relevant to keywords of the expanded search input, and relevant test failure records are prioritized using a relationship strength score based, at least in part, on relationship of the test failure records to the expanded search input. The facility detects test failure records from the searching having a relationship strength score higher than a threshold value, and analyzes the detected test failure records to identify potential test repair actions. Based on the analyzing, the facility generates a solution priority list including one or more suggested test repair actions for the test failure, and provides the solution priority list to the user.

Cognitive test advisor facility for identifying test repair actions

A method comprises computing an articulation rating associated with an input text, generating a modified subtext to modify a candidate subtext of the input text, generating an enhanced text that includes the modified subtext, computing an articulation rating associated with the enhanced text, and, based on the articulation rating associated with the enhanced text improving over the articulation rating associated with the input text, outputting the enhanced text. The method can include computing a meaning and/or relevance score associated with a candidate subtext of the input text, computing a meaning and/or relevance score associated with a modified subtext, and, based on one or both of the second scores improving over the respective first scores, generating the enhanced text to include the modified text. A word processing system, comprising a computerized input text, computerized attributes, and a computer-implemented word processor, can perform operations of the method.

Cognitive word processing

Systems and methods for content based routing are provided. Aspects include receiving, by a processor, a request to save a file. Analyzing, by the processor, data associated with the file. Determining one or more file save locations for the file based on a feature vector, generated by a machine learning model, comprising a plurality of features extracted from the data associated with the file and presenting the one or more file save locations to a user.

Content based recommendations of file system save locations

Methods, systems, and computer program products for placement of objects in a loading space are provided. Aspects include determining item data for a plurality of items, determining space data for a loading space, generating an optimized configuration for placement of the plurality of items in the loading space comprising determining one or more geometric representations for each item in the plurality of items, wherein the one or more geometric representations for each item are coupled together, determining, by an algorithm, a location in the loading space for each item in the plurality of items based on the one or more geometric representations for each item and the space data, and displaying, by a display, the optimized configuration comprising an image of each item of the plurality of items overlaid on corresponding locations in an image of the loading space.

Floor planning and placement of objects in a loading space

The embodiments relate to dynamic server drawer build plan management. Output data associated with a data source is received. A drawer build plan list and one or more inventory scenarios are generated based on the received data. At least one build plan is chosen from the build plan list, and one or more drawer design configurations are compared to at least one inventory scenario to produce a set of eligible drawer design configurations. A drawer design configuration is selected from the set for one or more respective drawer build plans based on one or more constraints. A connection is established to a computer, and information associated with each selected drawer design configuration is transmitted to the second computer.

Dynamic planning and configuration based on inconsistent supply

Goods/service recommendation that includes: accessing a consumer's computerized calendar to retrieve consumer calendar data and identifying an event in the calendar data; removing calendar data that is restricted data; determining first key words of the event representative of the event; identifying a target from the event; determining second key words of the target being descriptive of the target; determining third key words of the calendar data being descriptive of the consumer; combining the first, second and third key words; determining similarities and relationships between the target and the consumer to find groups of key words that form the strongest similarities and relationships; identifying the group of key words as having the strongest similarities and relationships between the target and the consumer; using the group of key words having the strongest similarities to identify a recommendation for providing a goods/service to the target; and providing the recommendation to the consumer.

Goods/service recommendation with data security

Embodiments of the present invention provide a computer-implemented method for generating test plans based on test failure root causes or symptoms. The method generates a heat map of manufacturing process test failures based at least in part on historical test failures and prior repair actions. A database is searched for test failure records that are relevant to user input. Relevant test failure records are prioritized via an index score that is assigned based at least in part on proximity of the relevant test failure records to the user input. Failure records that have an index score that is higher than a threshold value are detected. Unstructured text data of each of the detected failure records is analyzed to identify relevant keywords and relevancy rates. A test solution priority list is displayed, via a GUI, in a window or view that is separate from the heat map.

Cognitive manufacturing systems test repair action

Growing community of inventors

Beacon, NY, United States of America

Chanchal Saha