The gephi-graphs-deezer-project's intro from martamanevska

martamanevska / gephi-graphs-deezer-project Goto Github PK

View Code? Open in Web Editor NEW

By using Gephi’s modularity class feature, top 3 communities in subscription model were identified. No gender-related communities were detected. Based on centralities and hub measures there was found a list of users with efficient way transmitting information in the network, these top 100 users are recommended to target.

gephi-graphs-deezer-project's Introduction

About the dataset:

A social network of Deezer users which was collected from the public API in March 2020. Nodes are Deezer users from European countries and edges are mutual follower relationships between them. The vertex features are extracted based on the artists liked by the users. The task related to the graph is binary node classification - one has to predict the gender of users. This target feature was derived from the name field for each user.

Statistics:

Nodes 28,281
Edges 92,752
Density 0.0001
Transitivity 0.0959

Citing:

@misc{rozemberczki2020characteristic,
      title={Characteristic Functions on Graphs: Birds of a Feather, from Statistical Descriptors to Parametric Models},
      author={Benedek Rozemberczki and Rik Sarkar},
      year={2020},
      eprint={2005.07959},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

FEATHER paper:

https://arxiv.org/abs/2005.07959

FEATHER project:

https://github.com/benedekrozemberczki/FEATHER

Recommend Projects