In spite of furnishing high quality of education, demand on prognosticating pupil academic performance become more critical to ameliorate the quality and aiding scholars to achieve a great performance in their studies. The lack of having an effective and accurate vaticination model is one of the major issues today. The aim of this paper is to review current exploration conditioning related to academic analytics fastening on prognosticating pupil academic performance using various data mining techniques. Various data mining solutions have been proposed by former experimenters to develop the performance model using variety of scholar’s data, ways, algorithms and tools. Classification, regression, and clustering are just a few of the learning tasks that are related to the predictive modeling used to forecast student performance. Numerous variables have been picked and evaluated to determine the most important characteristics to do prediction in order to create the best prediction model. The ability to forecast performance accurately will be useful in guiding students' learning and helping them to avoid receiving low grades. Good input data and variables, a suitable predictive method, and a strong and reliable prediction model are all necessary in order to produce an effective predictive model.