I. INTRODUCTIONMost of the students have to do various projects in their curriculum. For number of subjects and fields, they need to select many project topics. These projects topics are required to have research papers associated with it. These papers give some idea to students for design, analysis, implementation and testing. Sometimes, they just have idea about domain in which they want to work. Due to large number of dataset on research paper's website, it is very time consuming task to find research papers which have work on similar problem with different technique to solve that problem. With 'Project Advisor', students can select field of their domain of interest and list of problems in that domain with list of related research papers will be shown as an output to students. This system is split into four phases-scrapping, classification, clustering, natural language processing. In the first phase of our system crawl through the web pages of research papers website to scrap title, abstract and various types of keywords. In the classification phase, it classifies papers into preselected four domains. The clusters of research papers with similar topics are found. In the final phase, a problem statement is generated for each cluster based on title and keywords. The input data is called training data and has a known label or result such as health-care or security at a time. A model is prepared through a training process where it is required to make predictions and is corrected when those predictions are wrong. Thus the system is trained till the model attains a required accuracy on the training data.