%0 Conference Paper %1 kdml23 %A Iqbal, Aftab %A Karnstedt., Marcel %B Proceedings of LWA2010 - Workshop-Woche: Lernen, Wissen & Adaptivitaet %C Kassel, Germany %D 2010 %E Atzmüller, Martin %E Benz, Dominik %E Hotho, Andreas %E Stumme, Gerd %K detection developer duplicate graph-based identification mining repositories room:0446 session:kdml2 sna software workshop:kdml %T On the Need of Graph Support for Developer Identification in Software Repositories %U http://www.kde.cs.uni-kassel.de/conf/lwa10/papers/kdml23.pdf %X Software repositories from open-source projects provide a rich source of information for a wide range of tasks. However, one issue to overcome in order to make this information useful is the accurate identification of developers. This is a particular challenge, as developers usually use different IDs in different repositories of one project, but usually there is no kind of dictionary or similar available to map the different IDs to real-world persons. Often, they even use different IDs in the same repository. We show that the few methods suggested so far are not always appropriate to overcome this problem. Further, we highlight related techniques from other areas and discuss how they can be applied in this context. We particularly focus on the idea of applying graph-based methods and argue for the benefits we expect from that.

@inproceedings{kdml23, abstract = {Software repositories from open-source projects provide a rich source of information for a wide range of tasks. However, one issue to overcome in order to make this information useful is the accurate identification of developers. This is a particular challenge, as developers usually use different IDs in different repositories of one project, but usually there is no kind of dictionary or similar available to map the different IDs to real-world persons. Often, they even use different IDs in the same repository. We show that the few methods suggested so far are not always appropriate to overcome this problem. Further, we highlight related techniques from other areas and discuss how they can be applied in this context. We particularly focus on the idea of applying graph-based methods and argue for the benefits we expect from that.}, added-at = {2010-10-05T14:15:12.000+0200}, address = {Kassel, Germany}, author = {Iqbal, Aftab and Karnstedt., Marcel}, biburl = {https://www.bibsonomy.org/bibtex/2c76bddb932c6c8194c3d5447849aa28b/lwa2010}, booktitle = {Proceedings of LWA2010 - Workshop-Woche: Lernen, Wissen {\&} Adaptivitaet}, crossref = {lwa2010}, editor = {Atzmüller, Martin and Benz, Dominik and Hotho, Andreas and Stumme, Gerd}, interhash = {9a46965a49963c6c4e050506a45c559c}, intrahash = {c76bddb932c6c8194c3d5447849aa28b}, keywords = {detection developer duplicate graph-based identification mining repositories room:0446 session:kdml2 sna software workshop:kdml}, presentation_end = {2010-10-05 17:45:00}, presentation_start = {2010-10-05 17:35:00}, room = {0446}, session = {kdml2}, timestamp = {2010-10-05T14:15:13.000+0200}, title = {On the Need of Graph Support for Developer Identification in Software Repositories}, track = {kdml}, url = {http://www.kde.cs.uni-kassel.de/conf/lwa10/papers/kdml23.pdf}, year = 2010 }