Local Partition in Rich Graphs

TL;DR

This paper introduces AttriPart, a scalable algorithm that enhances local graph partitioning by integrating attribute data, leading to denser partitions and faster computation, with applications in predicting community evolution.

Contribution

The paper presents AttriPart, a novel local graph partitioning algorithm that incorporates attribute data, improving partition quality and computational efficiency.

Findings

AttriPart finds up to 1.6× denser local partitions.

It runs approximately 43× faster than traditional methods.

LocalForecasting improves prediction accuracy of community evolution.

Abstract

Local graph partitioning is a key graph mining tool that allows researchers to identify small groups of interrelated nodes (e.g. people) and their connective edges (e.g. interactions). Because local graph partitioning is primarily focused on the network structure of the graph (vertices and edges), it often fails to consider the additional information contained in the attributes. In this paper we propose---(i) a scalable algorithm to improve local graph partitioning by taking into account both the network structure of the graph and the attribute data and (ii) an application of the proposed local graph partitioning algorithm (AttriPart) to predict the evolution of local communities (LocalForecasting). Experimental results show that our proposed AttriPart algorithm finds up to 1.6$\times$ denser local partitions, while running approximately 43$\times$ faster than traditional local partitioning techniques (PageRank-Nibble). In addition, our LocalForecasting algorithm shows a significant improvement in the number of nodes and edges correctly predicted over baseline methods.