New computer-assisted interactive learning methods and devices like intelligent tutoring systems, simulations, and games have increased the possibility of collecting and analysing student data, discovering patterns and trends in that data, and developing and testing new hypotheses about how students learn. In this paper, data mining techniques like Naïve Bayes method, Random Forest method, J48, Support Vector Machine and C4.5 classifier have been discussed. Each algorithm has its own advantages and disadvantages. Decision tree technique do not execute well if the data has smooth boundaries. The Naive Bayesian classifier works with both continuous and discrete attributes and operates well for real time problems. The objective of this review paper is to identify the appropriate technology that could be used to for data mining the database of the computer assisted learning tools to predict the right carrier for the students through their responses and interactions. This paper has focused on the probability of constructing a classification model for identifying student talents. Numerous attributes are tested, and a number of them have been found powerful on the performance identification.