Machine-Learning Based Objective Function Selection for Community Detection

TL;DR

This paper introduces NECTAR-ML, a machine-learning extension of the NECTAR community detection algorithm, which automates objective function selection and significantly improves detection quality over existing methods.

Contribution

NECTAR-ML employs a machine-learning model to automate objective function selection in community detection, outperforming the original NECTAR and other state-of-the-art algorithms.

Findings

Successfully selected the correct objective function in 90% of cases

NECTAR-ML outperforms NECTAR in detection quality

Outperforms multi-objective evolutionary algorithms in community detection

Abstract

NECTAR, a Node-centric ovErlapping Community deTection AlgoRithm, presented in 2016 by Cohen et. al, chooses dynamically between two objective functions which function to optimize, based on the network on which it is invoked. This approach, as shown by Cohen et al., outperforms six state-of-the-art algorithms for overlapping community detection. In this work, we present NECTAR-ML, an extension of the NECTAR algorithm that uses a machine-learning based model for automating the selection of the objective function, trained and evaluated on a dataset of 15,755 synthetic and 7 real-world networks. Our analysis shows that in approximately 90% of the cases our model was able to successfully select the correct objective function. We conducted a competitive analysis of NECTAR and NECTAR-ML. NECTAR-ML was shown to significantly outperform NECTAR's ability to select the best objective function. We also conducted a competitive analysis of NECTAR-ML and two additional state-of-the-art multi-objective community detection algorithms. NECTAR-ML outperformed both algorithms in terms of average detection quality. Multiobjective EAs (MOEAs) are considered to be the most popular approach to solve MOP and the fact that NECTAR-ML significantly outperforms them demonstrates the effectiveness of ML-based objective function selection.