Using Neuro-Fuzzy Technique to Classify and Predict Electrical Engineering Students’ Achievement Upon Graduation Based on Mathematics Competency

Abstract

This paper discusses the findings of a case study that uses neuro-fuzzy tool to classify and predict Electrical engineering students graduation achievement based on mathematics competency. In this study, achievement upon graduation and mathematics grades were classified as the key performance index. It's based on longitudinal progress and cross validation model on two mathematics subjects, semesters’ performance, and graduation achievement of electrical students. The outcomes indicated that there is a correlation between mathematics competency with electrical engineering performance, and it’s interesting to note that weak and satisfactory students in mathematics are not able to achieve first class upon graduation, and yet there is small percentage of excellent and good students in mathematics couldn’t graduate with high achievement. The findings conclude that the combination of statistical analysis and machine learning can help us to extract knowledge and enable university management to help low achievers at early stage. It’s hoped that the findings can help faculty management to review mathematics curriculum with respect to increasing range of engineering field.