US 11,941,364 B2
Context-driven analytics selection, routing, and management
Sushain Pandit, Austin, TX (US); Andrew T. Penrose, Castleknock (IE); Rory O Donnell, Dublin (IE); and Charles Daniel Wolfson, Austin, TX (US)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Sep. 1, 2021, as Appl. No. 17/463,658.
Prior Publication US 2023/0067828 A1, Mar. 2, 2023
Int. Cl. G06F 40/30 (2020.01); G06F 16/23 (2019.01); G06F 40/284 (2020.01)
CPC G06F 40/30 (2020.01) [G06F 16/2379 (2019.01); G06F 40/284 (2020.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving, by one or more processors, a user analytic request;
applying, by the one or more processors, natural language processing to the user analytic request;
associating, by the one or more processors, 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 stored in an analytics model registry by utilizing a plurality of domain-specific ontologies and a plurality of knowledge bases;
performing, by the one or more processors, a semantic search for one or more concepts of the first set of vectors and the second set of vectors to determine whether the one or more concepts are found in the one or more analytics models and data sources stored in the analytics model registry, wherein the step of performing further comprises:
narrowing, by the one or more processors, the one or more analytics models and data sources stored in the analytics model registry to the one or more analytics models and data sources that contain the one or more concepts;
tracking, by the one or more processors, a number of hops taken to discover the one or more concepts in the one or more analytics models and data sources stored in the analytics model registry;
calculating, by the one or more processors, a confidence metric for each of the one or more analytics models and data sources stored in the analytics model registry based on the number of hops taken to discover the one or more concepts in the one or more analytics models and data sources stored in the analytics model registry; and
ranking, by the one or more processors, the one or more analytics models and data sources stored in the analytics model registry based on the confidence metric calculated;
selecting, by the one or more processors, an analytics model from the one or more analytics models and data sources stored in the analytics model registry to process the user analytic request; and
executing, by the one or more processors, the analytics model for a time selected by a user, at a range selected by the user, and at a latitude position and a longitude position selected by the user.