US 12,254,013 B2
System and method for analysis of data from multiple data sources
Xiao-Ming Zhou, Zurich (CH); Tau Siang Pang, Zurich (CH); and Eric Bonabeau, Zurich (CH)
Assigned to Telepathy Labs, Inc., Tampa, FL (US)
Appl. No. 18/260,714
Filed by TELEPATHY LABS, INC., Tampa, FL (US)
PCT Filed Jan. 5, 2022, PCT No. PCT/US2022/011278
§ 371(c)(1), (2) Date Jul. 7, 2023,
PCT Pub. No. WO2022/150357, PCT Pub. Date Jul. 14, 2022.
Claims priority of provisional application 63/134,677, filed on Jan. 7, 2021.
Prior Publication US 2024/0378208 A1, Nov. 14, 2024
Int. Cl. G06F 16/2453 (2019.01); G06F 16/2455 (2019.01); G06F 16/2457 (2019.01)
CPC G06F 16/24575 (2019.01) [G06F 16/24539 (2019.01); G06F 16/24542 (2019.01); G06F 16/24561 (2019.01)] 14 Claims
OG exemplary drawing
 
1. A computer-implemented method for analysis of data from multiple data sources, the computer-implemented method comprising:
receiving a query instance;
processing the query instance to generate one or more relational query instances based upon, at least in part, the query instance;
distributing the generated one or more relational query instances among one or more nodes;
processing each of the generated one or more relational query instances by at least one node of one or more nodes to extract data from at least one data source of multiple data sources in communication therewith corresponding to the one or more relational query instances;
generating one or more insights based on the extracted data;
storing the one or more insights generated in a relevant cache to be retrieved therefrom;
identifying entities in the query instance and attributes associated therewith;
retrieving historic data from the at least one data source of the multiple data sources related to one or more of the entities and attributes identified;
training at least one model based on the historic data retrieved;
associating tags with the at least one trained model indicative of corresponding one or more of the entities and attributes; and
storing the at least one trained model with associated tags in at least one data source of the multiple data sources, to be retrieved therefrom.