Network Semantic Segmentation with Application to GitHub

TL;DR

This paper introduces network semantic segmentation for social network analysis, specifically applied to GitHub, by combining node attributes and network connections to classify users into meaningful topics.

Contribution

It proposes a novel approach that integrates network connections and attributes for semantic segmentation, enhancing community analysis in social coding networks.

Findings

Semantic segmentation improves community topic coherence

Method outperforms traditional community detection algorithms

Enhanced understanding of user interests in GitHub networks

Abstract

In this paper we introduce the concept of network semantic segmentation for social network analysis. We consider the GitHub social coding network which has been a center of attention for both researchers and software developers. Network semantic segmentation describes the process of associating each user with a class label such as a topic of interest. We augment node attributes with network significant connections and then employ machine learning approaches to cluster the users. We compare the results with a network segmentation performed using community detection algorithms and one executed by clustering with node attributes. Results are compared in terms of community diversity within the semantic segments along with topic