Multi-source Neural Topic Modeling in Multi-view Embedding Spaces

TL;DR

This paper introduces a neural topic modeling framework that leverages multi-view, multi-source embeddings, including pretrained topic and word embeddings, to enhance topic quality and handle polysemy across diverse document collections.

Contribution

It proposes a novel multi-source, multi-view embedding approach for neural topic modeling, improving performance on various datasets by integrating multiple pretrained embeddings.

Findings

Achieved state-of-the-art results on multiple datasets.

Improved topic coherence and perplexity scores.

Effective handling of polysemy and data sparsity.

Abstract

Though word embeddings and topics are complementary representations, several past works have only used pretrained word embeddings in (neural) topic modeling to address data sparsity in short-text or small collection of documents. This work presents a novel neural topic modeling framework using multi-view embedding spaces: (1) pretrained topic-embeddings, and (2) pretrained word-embeddings (context insensitive from Glove and context-sensitive from BERT models) jointly from one or many sources to improve topic quality and better deal with polysemy. In doing so, we first build respective pools of pretrained topic (i.e., TopicPool) and word embeddings (i.e., WordPool). We then identify one or more relevant source domain(s) and transfer knowledge to guide meaningful learning in the sparse target domain. Within neural topic modeling, we quantify the quality of topics and document representations via generalization (perplexity), interpretability (topic coherence) and information retrieval (IR) using short-text, long-text, small and large document collections from news and medical domains. Introducing the multi-source multi-view embedding spaces, we have shown state-of-the-art neural topic modeling using 6 source (high-resource) and 5 target (low-resource) corpora.