	US 12,014,140 B2
	Utilizing machine learning and natural language processing to determine mappings between work items of various tools
Koushik M. Vijayaraghavan, Chennai (IN); Niju Prabha, Bangalore (IN); Rajesh Nagarajan, Chennai (IN); Sarvesh Madhusudan Damle, Thane (IN); Aditi Kulkarni, Bengaluru (IN); Jayashri Sridevi, Chennai (IN); Rashmi Jhawar, Gurugram (IN); Prajwal Patrick Dsilva, Mumbai (IN); Soumya Swarup Jena, Bhubaneswar (IN); Dhiviya Dhanasekaran, Chennai (IN); Meiyarasu Selvam, Thevur (IN); and Girish Hulkoti, Bangalore (IN)
Assigned to Accenture Global Solutions Limited, Dublin (IE)
Filed by Accenture Global Solutions Limited, Dublin (IE)
Filed on May 3, 2021, as Appl. No. 17/306,629.
Prior Publication US 2022/0350967 A1, Nov. 3, 2022
Int. Cl. G06F 40/279 (2020.01); G06F 8/70 (2018.01); G06F 9/445 (2018.01); G06F 9/455 (2018.01); G06F 11/07 (2006.01); G06F 40/166 (2020.01); G06F 40/247 (2020.01); G06N 20/00 (2019.01)

CPC G06F 40/279 (2020.01) [G06F 8/70 (2013.01); G06F 40/166 (2020.01); G06F 40/247 (2020.01); G06N 20/00 (2019.01)]

20 Claims

1. A method, comprising:

receiving, by a device, work item data identifying work items associated with one or more of:

a user story associated with a project,

a defect associated with the project,

a test case associated with the project,

a requirement from a different tool of the project,

a first identifier indicating whether the requirement is satisfied, or

a second identifier indicating whether the requirement affects another requirement;

performing, by the device, data cleansing to remove and/or modify particular words from the work item data and to generate cleansed work item data;

performing, by the device, natural language processing on the cleansed work item data to identify synonyms for words in the cleansed work item data;

replacing, by the device, abbreviations in the cleansed work item data with full form text to generate final work item data;

identifying, by the device, keywords in the final work item data,

wherein the final work item data includes work items of the final work item data;

training, by the device, a machine learning model based on a neural network technique and using historical work item data associated with a plurality of historical projects, historical synonyms associated with the plurality of historical projects, and historical keywords associated with the plurality of historical projects to generate a trained machine learning model that is trained to identify mappings between the work items of the final work item data and to determine a confidence score for the mappings,

wherein the machine learning model includes a fuzzy model to identify the mappings;

processing, by the device, the final work item data, the synonyms, and the keywords, with the trained machine learning model, to identify the mappings and the confidence score,

wherein the confidence score, for a plurality of the work items, provides a measure of similarity between the plurality of the work items;

retraining, by the device, the trained machine learning model using the neural network technique and based on the mappings and the confidence score; and

performing, by the device, one or more actions based on the mappings and the confidence score for the mappings.