An approach for increasing educational performance using

Today, the ability to monitor and storage of large amounts of data has been provided with the progress of science and technology tools. There is a necessary need to search the data and extract useful knowledge among the ever-increasing data. Data analysis is automatic searching of large data sources to find patterns and dependencies which cannot be understood with simple statistical analyzes.Education is an area that is required to apply these tools for extensive data analysis and predictive modeling using new computational methods. The purposes of this research is indicating the role and the scope of predictive data analysis and propose a framework for evaluating and predicting students academic status with the operation of data analysis models. In this paper we study 11670 questionnaires that are distributed among five university students in Qom region in order to discover and highlight the important factors influencing over educational performance. As a result we can claim that the student who has higher rate of participation in class will receive a higher score.

Keywords:

Association rules , Data Mining , Higher Education , Student Retention , Rules’ extraction