Detect similarity in documents using GloVeTfIdf + RoBERTa + BERT!
You need glove.6b.300d.txt.gz for running glove model, and wikipedia document frequencies from Sentence Transformers.
https://public.ukp.informatik.tu-darmstadt.de/reimers/embeddings/
https://public.ukp.informatik.tu-darmstadt.de/reimers/embeddings/wikipedia_doc_frequencies.txt
Get dataset using python get_data.py.
Make sure u get tensorflow, scikit learn, numpy, etc.
Lot of thanks goes to UKP Lab, for providing the training of the models!