US 12,461,714 B2
System and method for generating a similarity matrix/score between intended requirements context data and source code context data
Sreemannarayana Balineni, Hyderabad (IN); Pushpendra Rathore, Bikaner (IN); Shaily Bhagwani, Raiper (IN); Priyatam Naravajhula, Hyderabad (IN); and KrishnaPrashanth Dharanikota, Hyderabad (IN)
Assigned to JPMORGAN CHASE BANK, N.A., New York, NY (US)
Filed by JPMorgan Chase Bank, N.A., New York, NY (US)
Filed on Jan. 14, 2022, as Appl. No. 17/648,023.
Claims priority of application No. 202111053306 (IN), filed on Nov. 19, 2021.
Prior Publication US 2023/0161562 A1, May 25, 2023
Int. Cl. G06F 8/30 (2018.01)
CPC G06F 8/30 (2013.01) 20 Claims
OG exemplary drawing
 
1. A method for developing an application by utilizing one or more processors and one or more memories, the method comprising:
converting requirements data into a first semantic context data, wherein the requirements data describes intended tasks required for developing an application, wherein the first semantic context data includes data defining technical context in developing the application, data validating intended tasks readiness for execution in developing the application, and data identifying duplicate intended tasks in developing the application;
converting the first semantic context data into a first semantic context vector;
accessing a database that stores source code corresponding to implementation of the intended tasks required for developing the application;
tokenizing the source code into sequence of actions;
training a machine learning model which receives a large dataset of word vectors as input and provides semantic context vectors as output, wherein an architecture of the machine learning model includes word embeddings section, hidden layers section, and dense layer section, and wherein data from the word embeddings section is input to the hidden layers section and data from the hidden layers section is input to the dense layer section, and wherein output from the machine learning model is input to a converter;
converting, by the converter, the source code into a second semantic context data;
converting, by the converter, the second semantic context data into a second semantic context vector, wherein the second semantic context vector describes an action performed by the source code;
comparing the first semantic context vector and the second semantic context vector;
automatically generating, in response to the comparing, a similarity score that indicates how much the source code and the requirements data are in line with each other;
determining, based on a result of the comparing, an amount of remaining source code necessary for completing the intended tasks required for the developing of the application;
recommending, based on the result of the comparing, an approach to further align the source code and the requirements data; and
executing development of the application when it is determined that the similarity score is equal to or more than a predetermined threshold value.