US 11,940,986 B1
Determining repair status information using unstructured textual repair data in response to natural language queries
Veysel Kocaman, Echt (NL); Julio Bonis Sanz, Algete (ES); and David Talby, Mercer Island, WA (US)
Assigned to John Snow Labs, Inc., Lewes, DE (US)
Filed by John Snow Labs, Inc., Lewes, DE (US)
Filed on Aug. 23, 2022, as Appl. No. 17/894,004.
Int. Cl. G06F 16/00 (2019.01); G06F 16/215 (2019.01); G06F 16/23 (2019.01); G06F 16/25 (2019.01); G06F 16/28 (2019.01)
CPC G06F 16/2365 (2019.01) [G06F 16/215 (2019.01); G06F 16/258 (2019.01); G06F 16/285 (2019.01)] 30 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
encoding, by one or more computing systems, content from multiple documents, wherein the multiple documents are from multiple sources, wherein the content from the multiple documents includes repair status information, wherein the repair status information includes structured data and unstructured data, wherein the repair status information further includes multiple types of repair and maintenance actions performed for a plurality of devices, and wherein the encoding of the content from the multiple documents includes:
converting, by the one or more computing systems, the content from the multiple documents into a common format, wherein the converting includes analyzing the content from the multiple documents to identity indications of the plurality of the devices, wherein the converting further includes analyzing the content from the multiple documents to identity indications of the multiple types of repair and maintenance actions, wherein the converting further includes analyzing the content from the multiple documents to identity indications of attributes associated with the indicated multiple types of repair and maintenance actions, and wherein the converting further includes modifying at least some of the unstructured data into a structured format used by at least some of the structured data;
separating, by the one or more computing systems, the converted content from the multiple documents into a plurality of data groupings, wherein each data grouping includes a subset of the repair status information about a respective one indicated repair or maintenance action performed for a respective indicated one of the plurality of devices, wherein the subset of the repair status information for each of at least some of the data groupings includes information about one or more attributes associated with the respective one indicated repair or maintenance action for that data grouping, and wherein the subset of the repair status information for each of the at least some data groupings further includes information from some of the unstructured data and from some of the structured data;
determining, by the one or more computing systems, one or more times associated with performing the respective one indicated repair or maintenance action for each data grouping, and associating the determined one or more times for each data grouping with that data grouping;
for each of the data groupings, validating data included in that data grouping, wherein the validating is performed by the one or more computing systems and includes verifying, for the respective indicated one device for that data grouping, consistency between first and second groups of data, wherein the first group of data is the subset of the repair status information for that data grouping and the associated determined one or more times for that data grouping, and wherein the second group of data is the subset of the repair status information for each of one or more other data groupings for the respective indicated one device for that data grouping, and the associated determined one or more times for that data grouping for each of the one or more other data groupings;
encoding, by the one or more computing systems, the plurality of data groupings to generate a plurality of encoded data groupings, wherein each of the encoded data groupings includes the subset of repair status information included in an associated one of plurality of data groupings and further includes additional supplemental information, wherein the additional supplemental information is from one or more additional documents and provides further information about the plurality of devices that is not repair or maintenance information, wherein generating of each of the plurality of encoded data groupings includes storing data in a graph database storage format, wherein the stored data includes data items of multiple types about the respective indicated one repair or maintenance action performed for the respective indicated one device for the associated one data grouping for that encoded data grouping; and
storing, by the one or more computing systems, the plurality of encoded data groupings in one or more graph databases in the graph database storage format;
encoding, by the one or more computing systems, a received query about an indicated type of repair or maintenance actions, wherein the received query is provided in a natural language format, and wherein the encoding of the received query includes generating a graph database query that represents semantic information of the received query;
determining, by the one or more computing systems, a response to the received query that provides information about the indicated type of the repair or maintenance actions performed for multiple devices of the plurality of devices, wherein the determining of the response includes:
identifying, by the one or more computing systems, multiple candidate encoded data groupings from the one or more graph databases that match the graph database query, wherein the multiple candidate encoded data groupings are a subset of the plurality of encoded data groupings; and
generating, by the one or more computing systems and using the multiple candidate encoded data groupings, the response to the received query, wherein the response includes information from some of the unstructured data and from some of the structured data, wherein the response further includes information about one or more attributes associated with the repair or maintenance actions performed for the multiple devices, wherein the response further includes information about timelines for the repair or maintenance actions performed for the multiple devices, and wherein the response further includes information about the verified consistency for the data groupings associated with the multiple candidate encoded data groupings; and
providing, by the one or more computing systems, the determined response to the received query, to initiate further repair or maintenance activities for at least one of the multiple devices, the initiated further repair or maintenance activities corresponding to the indicated type of the repair or maintenance actions.