Analysis and Prediction of Students’ Academic Performance and Employability Using Data Mining Techniques: A Research Travelogue

Abstract

Higher education institutions (HEIs) handle tons of data to analyze and generate the most relevant information. Data mining is considered a useful tool to extract knowledge to predict future educational trends in this process. Hence, such a method is significant to HEIs to understand and predict students’ employability and other critical academic elements. A comprehensive and systematic literature review was conducted to identify data mining techniques, algorithms, and the various data sets that will lead to the smart prediction and accuracy of student employability. The same method was used to determine the relationship between academic achievement and the employability of students. According to the research findings, the most frequently used data mining techniques for determining students' academic achievement and employability are Classification techniques, specifically the J48(C4.5) algorithm, the Naïve Bayes algorithm, and the CHAID Decision Tree algorithm. The most frequently used data sets or attributes for predicting students' academic performance and employability are their cumulative grade point average (CGPA), gender, technical, communication, problem-solving, analytical, critical thinking, and decision-making skills, extracurricular activities, and age, as well as psychomotor factors such as behavior and attendance and training/internship placement. Academic performance is the primary determinant of employability. The application of data mining techniques in academia has demonstrated its value in enhancing the performance of higher education institutions (HEIs). As a result, more research is urgently needed to ascertain the efficacy of the approaches, algorithms, and data sets identified as predictors of students' employability. Moreover, automated approaches should be utilized to ascertain their accuracy.