How is a data-driven approach better than random choice in label space division for multi-label classification?

TL;DR

This paper demonstrates that using community detection algorithms on label co-occurrence graphs significantly improves multi-label classification performance over random label partitioning, across multiple datasets and metrics.

Contribution

It introduces a novel approach of applying social network community detection methods to partition label space in multi-label classification, outperforming random partitioning methods.

Findings

Community detection methods outperform random partitioning in most cases.

Fastgreedy and walktrap on weighted graphs improve F1 scores by 85-92%.

Infomap on unweighted graphs improves Subset Accuracy and Jaccard similarity by about 90%.

Abstract

We propose using five data-driven community detection approaches from social networks to partition the label space for the task of multi-label classification as an alternative to random partitioning into equal subsets as performed by RAkELd: modularity-maximizing fastgreedy and leading eigenvector, infomap, walktrap and label propagation algorithms. We construct a label co-occurence graph (both weighted an unweighted versions) based on training data and perform community detection to partition the label set. We include Binary Relevance and Label Powerset classification methods for comparison. We use gini-index based Decision Trees as the base classifier. We compare educated approaches to label space divisions against random baselines on 12 benchmark data sets over five evaluation measures. We show that in almost all cases seven educated guess approaches are more likely to outperform RAkELd than otherwise in all measures, but Hamming Loss. We show that fastgreedy and walktrap community detection methods on weighted label co-occurence graphs are 85-92% more likely to yield better F1 scores than random partitioning. Infomap on the unweighted label co-occurence graphs is on average 90% of the times better than random paritioning in terms of Subset Accuracy and 89% when it comes to Jaccard similarity. Weighted fastgreedy is better on average than RAkELd when it comes to Hamming Loss.