Pluralistic Homophily

This ZIP file contains the datasets and results associated with the research on Pluralistic Homophily. The included data and results can be processed with code available at this Github repository.

File Contents:

This ZIP includes datasets from 9 networks that were the primary sources for the research:

     - StackOverflow     - DBLP     - Amazon     - LiveJournal     - YouTube     - Orkut     - Protein-Protein Interaction     - Drug-Drug Interaction     - C. elegans

Additionally, an example dataset based on the definition of a bee is included.   Each dataset consists of:

A network file in NCol format.

Files containing the ground-truth communities and those detected using the HLC algorithm (Anh et al.).

The NCol files were generated from datasets available at SNAP and BioSnap and were pre-processed for use in the experiments.

Results Files: The ZIP file also includes the files with the results of the research:

- Files in CSV format containing the calculations of pluralistic homophily at both the network and local levels for each of the mentioned networks.   

- These results include all metrics associated with the research and allow for the replication of the figures and tables presented in the paper.