rochan-a/topicmodeling

TopicModel

This repository contains scripts to train Topic Models on Hotel Reviews.
This work is done under the guidance of @Anupam Mediratta as an Intern.

Repo Structure

Each folder is named according to the training method or library used.

Folder	Description
VanillaLDA	Gensim LDA implementation
Mallet	MAchine Learning for LanguagE Toolkit
lda2vec Original, lda2vec used	Mix the best parts of word2vec and LDA into a single framework
depriciated	Gensim LDA and LSA
lda&w2vec	Incomplete custom training algorithm
postprocess	Scripts to process output from MALLET and Gensim LDAMallet
conditional_prob	Scripts to generate a conditional probability matrix