In this project, a large number of PDF files of research papers have to be analyzed using natural language processing techniques. The goals of the project involve providing insights into the semantic connections between documents, analyze and cluster the main research topics, provide a graphical overview and allow a user to quickly search the corpus for keywords and short sentences. The clustering can be done on the basis of keywords, and the semantic analysis between the documents will be done by the method where the words that are contained in the documents which have similar meanings occur together. For example, Latent Symantic Analysis can be used in this context. The results have to be presented through a web user interface, using the Dash+Plotly libraries. The web interface will help the users to search the corpus using short contextual sentences or keywords. We will also develop the graphical view of the embeddings of few keywords using an appropriate algorithm (for eg. t-SNE).