Your Extreme Multi-label Classifier is Secretly a Hierarchical Text Classifier for Free

TL;DR

This paper reveals that state-of-the-art extreme multi-label text classifiers inherently function as hierarchical classifiers, and evaluates their performance across traditional hierarchical and extreme label datasets.

Contribution

The study demonstrates that XML models effectively serve as hierarchical classifiers and highlights the limitations of HTC models on large-scale XML datasets.

Findings

XML models perform well as hierarchical classifiers.

HTC models struggle with large label sets.

Multiple metrics are necessary for fair evaluation.

Abstract

Assigning a set of labels to a given text is a classification problem with many real-world applications, such as recommender systems. Two separate research streams address this issue. Hierarchical Text Classification (HTC) focuses on datasets with label pools of hundreds of entries, accompanied by a semantic label hierarchy. In contrast, eXtreme Multi-Label Text Classification (XML) considers very large sets of labels with up to millions of entries but without an explicit hierarchy. In XML methods, it is common to construct an artificial hierarchy in order to deal with the large label space before or during the training process. Here, we investigate how state-of-the-art HTC models perform when trained and tested on XML datasets and vice versa using three benchmark datasets from each of the two streams. Our results demonstrate that XML models, with their internally constructed hierarchy, are very effective HTC models. HTC models, on the other hand, are not equipped to handle the sheer label set size of XML datasets and achieve poor transfer results. We further argue that for a fair comparison in HTC and XML, more than one metric like F1 should be used but complemented with P@k and R-Precision.