US 12,001,485 B2
Techniques for associating domain-specific terms for information discovery
Laurent Boué, Petah Tikva (IL); and Kiran Rama, Bangalore (IN)
Assigned to Microsoft Technology Licensing, LLP, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on May 27, 2022, as Appl. No. 17/826,927.
Prior Publication US 2023/0385340 A1, Nov. 30, 2023
Int. Cl. G06F 17/00 (2019.01); G06F 16/903 (2019.01); G06N 3/08 (2023.01)
CPC G06F 16/90344 (2019.01) [G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for information discovery using domain-specific term matching, comprising:
obtaining, for a search session related to a domain, a set of search strings used for searching the domain for information discovery during the search session;
identifying, for the search session, a string of the set of search strings as a search string hit, based on selection of a search result returned for the string to end the search session, and remaining strings in the set of search strings as search string misses;
correlating, into a set of domain-specific search string misses, the remaining strings in the set of search strings with additional remaining search strings from one or more other search session related to the domain that have a same string identified as the search string hit; and
providing the set of domain-specific search string misses along with the search string hit to a machine learning (ML) model that provides the search string hit as output given an input of at least one search string miss of the set of domain-specific search string misses or an input string having a certain distance from the at least one search string miss that is computed using a string similarity algorithm.