/LDA_from_scratch

We implement the Latent Dirichlet allocation(LDA) from scratch using python

Primary LanguageJupyter NotebookMIT LicenseMIT

LDA_from_scratch

We implement the Latent Dirichlet Allocation(LDA) from scratch using python, and compare our implementment with off the shelf ldamodel in gensim package.

We do the simulation test and real data test to verify the correctness of our implementation.

LDA_from_scratch.ipynb shows our implementation and LDA_from_gensim.ipynb is the off the shelf LDA package.

The real data set is abcnews-date-text.csv. The simulation dataset is simulation data.xlsx.

The .html is some visualization results from the test.