Proposal Software Predictor Based on a Probabilistic Model

In previous work we used data collected from studies of school career paths of graduates of Computer Science Faculty in Mazatlan city (FIM), of the Autonomous University of Sinaloa (UAS) in Mexico, to analyze and model those data with mathematical tools and discover information that allows an efficient and effective selection of candidates to enter university, through a prediction of the academic performance, decreasing the dropout rate and increasing the rate of discharge. Achieving an optimum selection of the students that get into the various options of careers that provides higher education in Mexico is one of the purposes of this study. Find the factors that influence the academic success of students and recent graduates is the main object of our research. The mathematical method of linear regression was applied through the SPSS software to analyze variables that influence the academic performance, among the factors studied are the gender, average high school, academic degree of the parents, income, marital status, previous High School score in the selection examination for admission to the university (Ceneval), among others, and discarding, according to the mathematical analyzes, those that do not affect the performance of the students. Resulting the average high school, gender, score in Ceneval and scores in Ceneval module in mathematics the most important variables. In this paper we present our mathematical model predictor of academic performance as the first step in order to develop a predictor software based in our mathematical model.