Automatic Generation of Topic Labels

TL;DR

This paper introduces a neural sequence-to-sequence model for automatic topic label generation, overcoming limitations of extractive methods by producing more flexible and human-like labels, trained on a large synthetic dataset.

Contribution

It presents a novel neural approach for topic labeling that generates descriptive labels without relying on a restricted candidate set, trained on a new synthetic dataset.

Findings

The neural model outperforms extractive methods in human evaluations.

Generated labels are more descriptive and relevant according to human ratings.

The approach demonstrates effectiveness across diverse topics.

Abstract

Topic modelling is a popular unsupervised method for identifying the underlying themes in document collections that has many applications in information retrieval. A topic is usually represented by a list of terms ranked by their probability but, since these can be difficult to interpret, various approaches have been developed to assign descriptive labels to topics. Previous work on the automatic assignment of labels to topics has relied on a two-stage approach: (1) candidate labels are retrieved from a large pool (e.g. Wikipedia article titles); and then (2) re-ranked based on their semantic similarity to the topic terms. However, these extractive approaches can only assign candidate labels from a restricted set that may not include any suitable ones. This paper proposes using a sequence-to-sequence neural-based approach to generate labels that does not suffer from this limitation. The model is trained over a new large synthetic dataset created using distant supervision. The method is evaluated by comparing the labels it generates to ones rated by humans.