The Implementation of Data Mining For Determining Higher Education Students‘ Performance
ABSTRACT
Advancements in communication and database technologies have revolutionized data handling for organizations, enabling them to efficiently collect, store, and manipulate vast amounts of information. Within the realm of higher education, understanding students’ behavior is a critical concern for academic institutions. Effective analysis and management of this growing data volume can facilitate informed decision-making processes.
The main objective of this study is to develop a predictive model that assesses higher education students’ performance through the application of data mining techniques. Following the six-step Hybrid methodology of the Knowledge Discovery Process model, the study encompasses understanding the problem domain, comprehending the data, data preparation, data mining, evaluating discovered knowledge, and utilizing the findings to achieve the set goal.
To explore the factors influencing higher education student performance, the study utilized data collected between 2006 and 2009. After data preparation involving thorough data cleaning, the study employed various classification algorithms, namely J48 Decision tree, PART Rule induction, Naïve Bayes, Logistic regression, Support Vector Machines, and Multilayer Perception Neural Network. These algorithms were chosen due to their popularity in recent related works. The experiments were conducted on a dataset comprising 11,550 instances, 21 attributes, and one outcome variable, utilizing the open-source data mining tool WEKA 3.9.2. The data was split into training and test datasets using 10-fold cross-validation and 66% split test modes.
The study’s results revealed that the J48 Decision tree algorithm achieved the highest classification accuracy, registering an impressive 97.84%. These findings are both interesting and promising, motivating the development of a predictive model for higher institution students’ performance. Key factors affecting student performance were identified, including the field of study, the number of common courses per semester, total courses per semester, academic year, financial source, and the number of supportive courses per semester.
In demonstrating the use of knowledge extracted through data mining, this study serves as a valuable example. For future endeavors, the integration of data mining with a knowledge system is recommended to design an intelligent system that can enhance decision-making processes in the academic setting.
The Implementation of Data Mining For Determining Higher Education Students‘ Performance, GET MORE COMPUTER SCIENCE PROJECT TOPICS AND MATERIALS