An automated approach to finding semantically related customer requests in the Jira bug tracking system

Kovalev A. D., Nikiforov I. V., Drobintsev P. D.

The work is devoted to research in the field of software maintenance phase automation. An automated approach to solving customer requests is proposed, which consists in using the Doc2Vec algorithm to search for semantically similar solved requests, as well as to find competent software engineers in the Jira issue tracking system. The proposed approach is implemented in a software tool, which allows to reduce the labor intensity of the maintenance stage by 12%. The results compare the manual and automated approach to analyzing customer requests in the process of software product support.

Keywords:software maintenance, automation, Doc2Vec, machine learning.