Mining twitter data for a more responsive software engineering process

Document Type

Conference Proceeding

Publication Date

6-30-2017

Abstract

Twitter has created an unprecedented opportunityfor software developers to monitor the opinions of large populationsof end-users of their software. However, automaticallyclassifying useful tweets is not a trivial task. Challenges stem fromthe scale of the data available, its unique format, diverse nature, and high percentage of spam. To overcome these challenges, thisextended abstract introduces a three-fold procedure that is aimedat leveraging Twitter as a main source of technical feedbackthat software developers can benefit from. The main objective isto enable a more responsive, interactive, and adaptive softwareengineering process. Our analysis is conducted using a dataset oftweets collected from the Twitter feeds of three software systems. Our results provide an initial proof of the technical value ofsoftware-relevant tweets and uncover several challenges to bepursued in our future work.

Publication Source (Journal or Book title)

Proceedings - 2017 IEEE/ACM 39th International Conference on Software Engineering Companion, ICSE-C 2017

First Page

280

Last Page

282

This document is currently not available here.

Share

COinS