Mining Java packages for developer profiles: An exploratory study

Jasmin Ramadani; Stefan Wagner

Mining Java packages for developer profiles: An exploratory study

Universität Stuttgart

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Not all developers have the same degree of knowledge of all parts of a software system. For allocating new task expertise, it would be interesting to have different developer profiles explicit. The state of the practice is to find out manually who might be most experienced in a certain area. A clear understanding how to automate this analysis is missing. Our goal is to explore to what degree the analysis of couplings of packages can be useful for this automation. Our analysis approach uses the idea that packages reflect the organization of the source code into basic functionalities. We use data mining on the version history to identify the sets of the packages that were most frequently changed together in different occasions. We present a case study where we analyze three open-source software systems to define developer expertise profiles based on the aggregation of the packages. Our results identify different developer profiles. They can be especially useful in analyzing projects with a larger number of developers saving time and effort by limiting the data sets to be investigated to find relevant software changes.

Original language	English
Title of host publication	Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband
Editors	Bernhard Mitschang, Norbert Ritter, Holger Schwarz, Meike Klettke, Andreas Thor, Oliver Kopp, Matthias Wieland
Publisher	Gesellschaft fur Informatik (GI)
Pages	143-152
Number of pages	10
ISBN (Electronic)	9783885796602
State	Published - 2017
Externally published	Yes
Event	Datenbanksysteme fur Business, Technologie und Web, BTW 2017 Workshopband - Database Systems for Business, Technology and Web, BTW 2017 Workshop - Stuttgart, Germany Duration: 6 Mar 2017 → 10 Mar 2017

Publication series

Name	Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)
Volume	266
ISSN (Print)	1617-5468
ISSN (Electronic)	2944-7682

Conference

Conference	Datenbanksysteme fur Business, Technologie und Web, BTW 2017 Workshopband - Database Systems for Business, Technology and Web, BTW 2017 Workshop
Country/Territory	Germany
City	Stuttgart
Period	6/03/17 → 10/03/17

Keywords

Developers expertise
Packages
Version history

Cite this

Ramadani, J., & Wagner, S. (2017). Mining Java packages for developer profiles: An exploratory study. In B. Mitschang, N. Ritter, H. Schwarz, M. Klettke, A. Thor, O. Kopp, & M. Wieland (Eds.), Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband (pp. 143-152). (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. 266). Gesellschaft fur Informatik (GI).

Ramadani, Jasmin ; Wagner, Stefan. / Mining Java packages for developer profiles : An exploratory study. Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband. editor / Bernhard Mitschang ; Norbert Ritter ; Holger Schwarz ; Meike Klettke ; Andreas Thor ; Oliver Kopp ; Matthias Wieland. Gesellschaft fur Informatik (GI), 2017. pp. 143-152 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

@inproceedings{8a104c9ae7544fe7913e7e1715cc7c6c,

title = "Mining Java packages for developer profiles: An exploratory study",

abstract = "Not all developers have the same degree of knowledge of all parts of a software system. For allocating new task expertise, it would be interesting to have different developer profiles explicit. The state of the practice is to find out manually who might be most experienced in a certain area. A clear understanding how to automate this analysis is missing. Our goal is to explore to what degree the analysis of couplings of packages can be useful for this automation. Our analysis approach uses the idea that packages reflect the organization of the source code into basic functionalities. We use data mining on the version history to identify the sets of the packages that were most frequently changed together in different occasions. We present a case study where we analyze three open-source software systems to define developer expertise profiles based on the aggregation of the packages. Our results identify different developer profiles. They can be especially useful in analyzing projects with a larger number of developers saving time and effort by limiting the data sets to be investigated to find relevant software changes.",

keywords = "Developers expertise, Packages, Version history",

author = "Jasmin Ramadani and Stefan Wagner",

note = "Publisher Copyright: {\textcopyright} 2017 Gesellschaft fur Informatik (GI). All rights reserved.; Datenbanksysteme fur Business, Technologie und Web, BTW 2017 Workshopband - Database Systems for Business, Technology and Web, BTW 2017 Workshop ; Conference date: 06-03-2017 Through 10-03-2017",

year = "2017",

language = "English",

series = "Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)",

publisher = "Gesellschaft fur Informatik (GI)",

pages = "143--152",

editor = "Bernhard Mitschang and Norbert Ritter and Holger Schwarz and Meike Klettke and Andreas Thor and Oliver Kopp and Matthias Wieland",

booktitle = "Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband",

}

Ramadani, J & Wagner, S 2017, Mining Java packages for developer profiles: An exploratory study. in B Mitschang, N Ritter, H Schwarz, M Klettke, A Thor, O Kopp & M Wieland (eds), Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI), vol. 266, Gesellschaft fur Informatik (GI), pp. 143-152, Datenbanksysteme fur Business, Technologie und Web, BTW 2017 Workshopband - Database Systems for Business, Technology and Web, BTW 2017 Workshop, Stuttgart, Germany, 6/03/17.

Mining Java packages for developer profiles: An exploratory study. / Ramadani, Jasmin; Wagner, Stefan.
Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband. ed. / Bernhard Mitschang; Norbert Ritter; Holger Schwarz; Meike Klettke; Andreas Thor; Oliver Kopp; Matthias Wieland. Gesellschaft fur Informatik (GI), 2017. p. 143-152 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. 266).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Mining Java packages for developer profiles

T2 - Datenbanksysteme fur Business, Technologie und Web, BTW 2017 Workshopband - Database Systems for Business, Technology and Web, BTW 2017 Workshop

AU - Ramadani, Jasmin

AU - Wagner, Stefan

PY - 2017

Y1 - 2017

N2 - Not all developers have the same degree of knowledge of all parts of a software system. For allocating new task expertise, it would be interesting to have different developer profiles explicit. The state of the practice is to find out manually who might be most experienced in a certain area. A clear understanding how to automate this analysis is missing. Our goal is to explore to what degree the analysis of couplings of packages can be useful for this automation. Our analysis approach uses the idea that packages reflect the organization of the source code into basic functionalities. We use data mining on the version history to identify the sets of the packages that were most frequently changed together in different occasions. We present a case study where we analyze three open-source software systems to define developer expertise profiles based on the aggregation of the packages. Our results identify different developer profiles. They can be especially useful in analyzing projects with a larger number of developers saving time and effort by limiting the data sets to be investigated to find relevant software changes.

AB - Not all developers have the same degree of knowledge of all parts of a software system. For allocating new task expertise, it would be interesting to have different developer profiles explicit. The state of the practice is to find out manually who might be most experienced in a certain area. A clear understanding how to automate this analysis is missing. Our goal is to explore to what degree the analysis of couplings of packages can be useful for this automation. Our analysis approach uses the idea that packages reflect the organization of the source code into basic functionalities. We use data mining on the version history to identify the sets of the packages that were most frequently changed together in different occasions. We present a case study where we analyze three open-source software systems to define developer expertise profiles based on the aggregation of the packages. Our results identify different developer profiles. They can be especially useful in analyzing projects with a larger number of developers saving time and effort by limiting the data sets to be investigated to find relevant software changes.

KW - Developers expertise

KW - Packages

KW - Version history

UR - http://www.scopus.com/inward/record.url?scp=85072782958&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85072782958

T3 - Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)

SP - 143

EP - 152

BT - Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband

A2 - Mitschang, Bernhard

A2 - Ritter, Norbert

A2 - Schwarz, Holger

A2 - Klettke, Meike

A2 - Thor, Andreas

A2 - Kopp, Oliver

A2 - Wieland, Matthias

PB - Gesellschaft fur Informatik (GI)

Y2 - 6 March 2017 through 10 March 2017

ER -

Ramadani J, Wagner S. Mining Java packages for developer profiles: An exploratory study. In Mitschang B, Ritter N, Schwarz H, Klettke M, Thor A, Kopp O, Wieland M, editors, Datenbanksysteme fur Business, Technologie und Web, BTW 2017 - Workshopband. Gesellschaft fur Informatik (GI). 2017. p. 143-152. (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

Mining Java packages for developer profiles: An exploratory study

Abstract

Publication series

Conference

Keywords

Other files and links

Fingerprint

Cite this