In olden days students exposed to teaching and learning was confined only to classroom. Students in the 21st century are connected well with lot of literature and learning from various open source platforms. Hence the students find it difficult to identify the right content for their learning. In this context data mining tools which are evolved in the past decade are of great use in filtering the large volumes of data and to pick the appropriate content for the student. In the previous years a considerable amount of time has been spent on student profiles and not on the factors responsible for the varied performances of the students. Data mining plays an important role in this field and can be used to mine relevant data for further optimization. This paper explains about how various techniques of educational data mining can be used to identify the student profiles and their behavior on the social media by taking many factors into consideration, to forecast the students’ performance and also identify the best suited curriculum structure for them, to understand the pitfalls in teaching-learning environment etc. All these predictions can be used to help the ‘at risk’ student’s category. The aim of the paper is to contribute to the literature on the application of data mining in the field of education by providing an overview of the various mining techniques that can be used, challenges faced in implementing the techniques and a comparative study of all the techniques. © 2016, Institute of Integrative Omics and Applied Biotechnology. All rights reserved.