FinancialNarrativeSummarization

Project based on Machine Learning, BERT, NLP. Implementing state-of-the-art models such as the pre-trained model Distil BERT(a variant of BERT) with K-Means, Text-Rank, Page-Rank, LSA, Graph-based models, Gensim and Sentence Ranking model where sentences selected using cosine distance similarity metric, TF-IDF scores, etc. Lastly, we have compared all these models with the SCE-SUMM model to evaluate the effectiveness of our models with the help of ROUGE scores. The TF-IDF summ model performs better than the other models, with the highest Rouge-L score of 46.2%