An Information-Theoretic Approach to Data Selection in Generative Topic Modeling

This repository is the implementation of the paper "An Information-Theoretic Approach to Data Selection in Generative Topic Modeling".

Getting Started

Install Python 3.11.9 in your local directory and install the requirements.

Workspace

Following directories should be created for our experiments.

• ./model_results/ -> storing the output of the experiment
• ./dataset/ -> storing the data used for the experiment

How to Run

Data Selection Algorithm

open src/document_selection_algorithm.ipynb.

To visualize the results

open src/visualization.ipynb.

Contact

Michael Evan Santoso - is0534is@ed.ritsumei.ac.jp