Research And Application Of Performance Prediction Model Based On Student Behavior

With the rapid development of information technology,information management system has replaced the traditional paper record mode,being more and more popular in the daily production and management among all walks of life.Since the 21 st century,major universities to accelerate the pace of information construction,build their own information center,use the information management system to store and manage educational data.After more than a decade of accumulation,the database of university information center stored a large number of student data,the past the value of these data is not taken seriously,often overlooked.In the context of the current global big data,these data began to enter the researcher's vision.How to extract useful information from these data to help students improve their learning and improve the level of teaching management in universities has also become an import subject in the field of data mining.Academic achievement is the most import and basic indicator of a student's learning situation.The analysis of academic performance is conducive to teacher to master the students' learning situation,targeted teaching counseling.And students behavior data is the students in the campus of learning methods and living habits of the reflection.This paper,use the data mining to mine and analyze the college students living habits,research potential patterns between students behavior and student achievement which has a strong correlation according to university student behavior data,and establish the forecast model through data mining.In order to use students' daily behavior to predict student achievement while establishing student counseling mechanisms to improve student achievement.Performance analysis and forecasting is a new research direction of educational data mining.Most of the existing research is based on the analysis of curriculum or historical achievements.A small number of behavior-based analyzes are also focused on the association rules mining between single behavior and achievement.Based on the previous studies,this paper does not depend on the students' personal information and past achievements,but rather the behavior data generated by students in daily campus activities,including consumption data,access control data and borrowing data with a lot of in-depth analysis to find the behavioral patterns behind student behavior data and factors related to student achievement.Select two grades behavioral data,to extract behavioral features,using data mining algorithms to build predictive models of student achievement,while the effects of different models to predict the results were compared to determine the best model to predict the results.Finally,the results of this study will be applied to the real student management system.This paper uses the data mining to predict the student achievement from a new angle,and makes some exploratory work for the application of data mining in the field of education.