There is an important requirement to classify unknown documents based upon their content for similarity checking. This project aims to take a dataset of research papers and classify them into subtopics by using the K Means Algorithm so that researchers can easily follow work which is closely related to their own field.