IR-Pubmed_QA

Stage 1 of my mini project using PubmedQA dataset contains three types of dataset namely PQAL-L,PQAU-U,PQAA. In that, I'm using PQAL(Labeled). Basic idea behind this project is that, a user ask the query,transform the query into vectors and using similiarity score we compute the answers for the given query. Method - Data Preprocessing, Feature extraction, TF-IDF, Cosine similiarity, Dimensionality reduction. Tools: NLP,JupyterNotebook,Pandas,EDA,ML