Self-supervised Topic Taxonomy Discovery in the Box Embedding Space

TL;DR

This paper introduces BoxTM, a novel box embedding-based model for hierarchical topic taxonomy discovery that effectively captures semantic scopes and asymmetric relations, resulting in higher quality topic hierarchies.

Contribution

The paper proposes a new box embedding approach for topic modeling that better captures hierarchical relations and semantic scopes compared to traditional Euclidean methods.

Findings

High-quality topic taxonomies learned by BoxTM

Effective inference of asymmetric hierarchical relations

Recursive clustering improves upper-level topic discovery

Abstract

Topic taxonomy discovery aims at uncovering topics of different abstraction levels and constructing hierarchical relations between them. Unfortunately, most of prior work can hardly model semantic scopes of words and topics by holding the Euclidean embedding space assumption. What's worse, they infer asymmetric hierarchical relations by symmetric distances between topic embeddings. As a result, existing methods suffer from problems of low-quality topics at high abstraction levels and inaccurate hierarchical relations. To alleviate these problems, this paper develops a Box embedding-based Topic Model (BoxTM) that maps words and topics into the box embedding space, where the asymmetric metric is defined to properly infer hierarchical relations among topics. Additionally, our BoxTM explicitly infers upper-level topics based on correlation between specific topics through recursive clustering on topic boxes. Finally, extensive experiments validate high-quality of the topic taxonomy learned by BoxTM.