Neural Topic Modeling

Testing of Neural Topic Modeling for Japanese articles

This repository intended to test performance of Neural Topic Model and Gensim LDA. A Pytorch implementation of Neural Topic Model and Gensim LDA test code are contained in this repository.

Dataset

livedoor ニュースコーパス / livedoor News Corpus
https://www.rondhuit.com/download.html#ldcc
CC BY-ND 2.1 JP
https://creativecommons.org/licenses/by-nd/2.1/jp/

Neural Topic Model

Reference

• yuewang-cuhk/TAKG https://github.com/yuewang-cuhk/TAKG
• ysmiao/nvdm https://github.com/ysmiao/nvdm/
• http://tdual.hatenablog.com/entry/2018/04/09/133000

memo

• yuewang-cuhk' s NTM implementation is partially used.
• tdual' s script is massively cited.
• janome tokenizer is used instead of Mecab.

GPU instance is recommended. If training is too slow, please check instance type of Google Colaboratory.

LDA

Reference

http://tdual.hatenablog.com/entry/2018/04/09/133000

Memo

• tdual' s LDA script is massively cited.
• janome tokenizer is used instead of Mecab.