US 12,488,301 B2
Method and system for automated scoring of development tasks by using artificial intelligence
Pranav Raman, Hoffman Estates, IL (US); Terrence Cook, Country Club Hills, IL (US); Nitin Challa, Aurora, IL (US); Grace Kim, Chicago, IL (US); Caleb McClatchey, Ellicott City, MD (US); and Venkatraghavan Kannan, Novi, MI (US)
Assigned to JPMORGAN CHASE BANK, N.A., New York, NY (US)
Filed by JPMorgan Chase Bank, N.A., New York, NY (US)
Filed on Oct. 12, 2023, as Appl. No. 18/379,402.
Prior Publication US 2025/0124389 A1, Apr. 17, 2025
Int. Cl. G06Q 10/0639 (2023.01); G06Q 10/0631 (2023.01); G06Q 10/10 (2023.01)
CPC G06Q 10/06393 (2013.01) [G06Q 10/063114 (2013.01); G06Q 10/103 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for facilitating automated scoring of software development tasks by using predictive analytics, the method being implemented by at least one processor, the method comprising:
receiving, by the at least one processor via a graphical user interface, at least one request, each of the at least one request including a search request for at least one task;
aggregating, by the at least one processor, raw data that corresponds to the at least one task, the raw data including information in a natural language format;
generating, by the at least one processor, at least one structured data set from the raw data based on at least one predetermined parameter;
determining, by the at least one processor using a first model, at least one task score for each of the at least one task based on the at least one structured data set;
training a second model based on output of the first model corresponding to distribution of probabilities generated by the first model;
determining, by the at least one processor using the second model, a confidence score for each of the at least one task score;
determining, by the at least one processor, whether the confidence score satisfies a predetermined threshold;
identifying, by the at least one processor, at least one set of adjacent scores based on an adjacency between each of the at least one task score, wherein the at least one set of adjacent scores is determined when the confidence score is less than the predetermined threshold;
training a third model based on the output of the first model;
determining, by the at least one processor using the third model, an adjacency probability value for each of the at least one set of adjacent scores;
computing, by the at least one processor, at least one predictive output for each of the at least one task based on the corresponding at least one task score, the corresponding adjacency probability value, and the corresponding confidence score;
transmitting, by the at least one processor via the graphical user interface, the at least one predictive output to a downstream application including a downstream development platform;
tracking the software development tasks based on the at least one predictive output to fulfill a goal of completing development of the software,
wherein the downstream application is a microservice application that relates to an event-driven architecture that includes a distributed data streaming platform for publishing, subscribing, storing, and processing of event streams in real time.