Data Mining Behavioral Transitions in Open Source Repositories

Author

Robinson, William N. ; Tianjie Deng

Author_Institution

Comput. Inf. Syst. Dept., Georgia State Univ., Atlanta, GA, USA

fYear

2015

fDate

5-8 Jan. 2015

Firstpage

5280

Lastpage

5289

Abstract

Open-source repository data can be automatically mined using sequence mining methods to provide high-level feedback on project status. GitHub.com projects are acquired, sequence-mined, clustered, and regressed to analyze project characteristics. Such results can be presented to project managers, as part of a display generated by an automated monitoring system. Such monitoring systems provide high-level feedback in real-time. This project is a preliminary step in a larger research project aimed at understanding and monitoring FLOSS projects using this process modeling approach.

Keywords

behavioural sciences computing; data mining; pattern clustering; project management; public domain software; regression analysis; FLOSS projects; GitHub.com projects; automated monitoring system; data clustering; data mining behavioral transitions; high-level feedback; open-source repository data; process modeling approach; project characteristics analysis; project managers; regression analysis; research project; sequence mining methods; Analytical models; Cognition; Data mining; Data models; Hidden Markov models; Monitoring; Software;

fLanguage

English

Publisher

ieee

Conference_Titel

System Sciences (HICSS), 2015 48th Hawaii International Conference on

Conference_Location

Kauai, HI

ISSN

1530-1605

Type

conf

DOI

10.1109/HICSS.2015.622

Filename

7070450