Zelko, Andreea Cristina (2024) Location inference using GitHub profiles. Master's Internship Report, Computing Science.
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
Thesis deposit from_ Andreea Cristina Zelko _ degree programme_ Computing Science.pdf Restricted to Registered users only Download (132kB) |
|
|
Text
mCS_2024_ZelkoAC.pdf Download (309kB) | Preview |
Abstract
One interesting focus within the software analytics research domain is understanding the impact of international collaboration on the productivity of a development team. In order for such research to be carried out, researchers need access to the locations of the team members. Often, this information is only shared by a small number of users, meaning that a lot of users can not be used in research due to insufficient data. The goal of this research project is to explore methods of obtaining locations which can be associated with the GitHub profile of a user. Being able to infer the location of users using only their profile would enable researchers to expand their candidate pool. This would result in more generalizable and more accurate findings. The name, bio and email present in the GitHub profile of a person are used to infer locations. The results of the inference show that names can easily be used to obtain a nationality, while the bio and email are harder to use since they are not always provided. It is concluded that inference methods are best used in combination with each other, so that multiple information sources get analysed.
| Item Type: | Thesis (Master's Internship Report) |
|---|---|
| Supervisor name: | Rastogi, A. and Capiluppi, A. |
| Degree programme: | Computing Science |
| Thesis type: | Master's Internship Report |
| Language: | English |
| Date Deposited: | 24 Jun 2024 09:11 |
| Last Modified: | 24 Jun 2024 12:05 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/32732 |
Actions (login required)
 |
View Item |