Toward Interpretable Topic Discovery via Anchored Correlation Explanation
This work addresses the challenge of incorporating informal expert knowledge into topic modeling for tasks like medical diagnosis, though it appears incremental as it builds on existing methods.
The paper tackled the problem of encoding expert knowledge for interpretable topic discovery by proposing Anchored CorEx, which combines the information bottleneck and Total Correlation Explanation to guide topic modeling with fuzzy expert input, resulting in more coherent and interpretable topics on two corpora.
Many predictive tasks, such as diagnosing a patient based on their medical chart, are ultimately defined by the decisions of human experts. Unfortunately, encoding experts' knowledge is often time consuming and expensive. We propose a simple way to use fuzzy and informal knowledge from experts to guide discovery of interpretable latent topics in text. The underlying intuition of our approach is that latent factors should be informative about both correlations in the data and a set of relevance variables specified by an expert. Mathematically, this approach is a combination of the information bottleneck and Total Correlation Explanation (CorEx). We give a preliminary evaluation of Anchored CorEx, showing that it produces more coherent and interpretable topics on two distinct corpora.