geometric-intelligence
Category: A1; Team name: GAAIMC; Dataset: Github (MUSAE)
Checklist
- [x] My pull request has a clear and explanatory title.
- [x] My pull request passes the Linting test.
- [x] I added appropriate unit tests and I made sure the code passes all unit tests. (refer to comment below)
- [x] My PR follows PEP8 guidelines. (refer to comment below)
- [x] My code is properly documented, using numpy docs conventions, and I made sure the documentation renders properly.
- [x] I linked to issues and PRs that are relevant to this PR.
Description
This pull request adds the Github (MUSAE) dataset published in [1] for TAG-DS Topological Deep Learning Challenge 2025: expanding the data landscape.
This dataset contains a large social network of GitHub developers which was collected from the public API in June 2019. Nodes are developers who have starred at least 10 repositories and edges are mutual follower relationships between them. The vertex features are extracted based on the location, repositories starred, employer and e-mail address. The task related to the graph is binary node classification - one has to predict whether the GitHub user is a web or a machine learning developer. This target feature was derived from the job title of each user. [2]
This dataset was shared in PyG [3], but the url to download it is broken, so we downloaded it from [2]. In [4] the features were truncated to a dimensionality of 128 using SVD. We added the dimensionality reduction as a data transformation and is performed as default for this dataset, however the complete data is kept, in case of choosing another kind of data transformation.
The same data transformation is used in PR's #214, #216 and #229
References:
[1] B. Rozemberczki, C. Allen and R. Sarkar. Multi-scale Attributed Node Embedding. 2019. [2] SNAP: Network Datasets: Github (MUSAE) [3] Github in PyG [4] B. Rozemberczki and R. Sarkar. Characteristic Functions on Graphs: Birds of a Feather, from Statistical Descriptors to Parametric Models. 2020.
