Research On The Correlation Between Daily Behavior And Performance Of College Students Based On Apriori

With the advent of the era of big data,the speed of informatization in the field of education continues to improve,and the process of digital campus in colleges and universities is also advancing steadily.While bringing convenience to teachers and students,college platforms and business systems store a large amount of data,a large part of which is generated by students during school days,these massive data resources provide indispensable data support for data mining analysis.It is of great practical significance to carry out individualized teaching management in colleges and universities,adopt different education modes for different students and improve the efficiency of schoolwork.Based on this,this paper collects student data from university academic affairs system,one-card system,library management system,academic affairs office grade management system,teaching management system and various business systems of the university.Based on the pre-processed data,a database of behavioral characteristics of college students is established,and the indicators are divided into three major sections of consumer behavior,living behavior,and learning behavior according to student behavior.By selecting the initial centroid and reducing the number of distance operations,the traditional K-means algorithm is optimized to improve the accuracy and operation speed of the algorithm,and the clustering of student behavior features is mined,and the clustering results are elaborated.This study improves the Apriori algorithm using both prepruning optimization and bi-directional search optimization to improve the disadvantages of the traditional Apriori algorithm of scanning the database too many times and having a large time overhead.Based on the comparison of the running times of the three algorithms tested under different support thresholds,it is found that the bidirectional search optimization algorithm outperforms the prepruning optimization algorithm and the traditional Apriori algorithm.Therefore,the improved algorithm based on bidirectional search is finally selected to conduct correlation mining analysis on students’ behavior characteristics and their grades.The clustered student behavior data information is organized,and association rules are mined between them and student grades.All association rules are screened according to the research content of this paper to get the relationship between student behavior patterns and their grades,and finally five effective association rules are mined.According to the mining results,it can provide decision-making direction for the educational and teaching management tasks of colleges and universities,which is conducive to educators to improve the quality of classroom teaching,accurately implement the work of teaching students in accordance with their aptitude,improve the overall quality of students,start the academic warning work of colleges and universities,and improve the efficiency of teaching,etc.It also provides some support for college administrators to formulate,improve and implement the corresponding policies.