Embedded Topic Models Enhanced by Wikification

TL;DR

This paper introduces an enhanced neural topic model that incorporates Wikipedia knowledge to recognize named entities, improving topic coherence and capturing temporal dynamics in document collections.

Contribution

It presents a novel method integrating Wikification into neural topic models, enabling better recognition of entities and temporal topic evolution.

Findings

Improved generalizability of topic models on news and dataset.

Enhanced recognition of named entities in topics.

Effective modeling of temporal topic development.

Abstract

Topic modeling analyzes a collection of documents to learn meaningful patterns of words. However, previous topic models consider only the spelling of words and do not take into consideration the homography of words. In this study, we incorporate the Wikipedia knowledge into a neural topic model to make it aware of named entities. We evaluate our method on two datasets, 1) news articles of \textit{New York Times} and 2) the AIDA-CoNLL dataset. Our experiments show that our method improves the performance of neural topic models in generalizability. Moreover, we analyze frequent terms in each topic and the temporal dependencies between topics to demonstrate that our entity-aware topic models can capture the time-series development of topics well.