Maryam Mollabagher - Faculty member of University of Applied Science & Technology, Tehran, Iran,
Mostafa Yousefi Tezerjan - Faculty member of University of Applied Science & Technology, Tehran, Iran,
Esrafil Ala - General Manager of the Bureau of Measurement and Testing, Tehran, Iran,

The aim of this study is to provide data analysis capabilities in the analysis of centralized test data at The University of Applied Sciences & Technology. Due to the wide range of data mining techniques and the variety of different types of data in the database, in order to achieve the objectives of data mining, the research process must be defined very accurately. Considering that this research was done in the university environment and through the information available in the university systems, so there was no problem in obtaining the information. In order to achieve the knowledge gained from the centralized exam, clustering of students, teachers, provinces participating in this exam, identifying the patterns available in The University of Applied Sciences & Technology to provide a model to predict students' scores in centralized exams in the next semester. ۱۹۲۰۷ student / course scores have been studied in ۸ subjects in ۲۸ educational centers that have been held in an associate's and bachelor's degree level and at the same time throughout the country. Using the feature selection method, the most effective ones were selected. Then, preprocessing data mining techniques including clustering.For this purpose, initial and supplementary statistics were first extracted and after clustering by K -means method, ۵ clusters were obtained, a model for predicting students' scores in the upcoming semester is presented in the focused exams courses.This prediction pattern can be effective in making the learning process more efficient in the academic system. Respectively the use of a decision -making tree model is also a model for predicting students' grades in the next semester in centralized exam courses.

Centralized Testing, Educational Data Mining, Clustering, Prediction