"Draw My Topics": Find Desired Topics fast from large scale of Corpus

TL;DR

The paper introduces 'Draw My Topics', a toolkit that quickly identifies relevant topics in large corpora by integrating social scientists' interests with topic models using vector space and entropy methods.

Contribution

It presents a novel toolkit combining vector space and entropy techniques to align topic modeling with social scientists' specific interests efficiently.

Findings

Effective in identifying desired topics from large corpora

Allows user adjustments for specific interests

Demonstrated on Chinese Diachronic People's Daily Corpus

Abstract

We develop the "Draw My Topics" toolkit, which provides a fast way to incorporate social scientists' interest into standard topic modelling. Instead of using raw corpus with primitive processing as input, an algorithm based on Vector Space Model and Conditional Entropy are used to connect social scientists' willingness and unsupervised topic models' output. Space for users' adjustment on specific corpus of their interest is also accommodated. We demonstrate the toolkit's use on the Diachronic People's Daily Corpus in Chinese.