<![CDATA[This research aims to analyze and predict student achievement using data mining techniques with the C4.5 and Naive Bayes methods. The data used includes various factors that affect students' academic performance, such as previous grades, attendance, and parents' income. The C4.5 method, which is a decision tree algorithm, is used to identify patterns in the data and make rule-based decisions. Meanwhile, Naive Bayes, which is a probabilistic classification technique, is used to calculate the probability of achievement based on the distribution of features. The C4.5 algorithm model showed excellent performance in classifying students into the categories of “Underachieving” and “Achieving,” with perfect accuracy and F1-Score for both classes. On the other hand, the Naive Bayes model showed less than optimal results, especially in recognizing “Outstanding” students. Although the Naive Bayes model managed to correctly predict all the “Underachieving” students, it failed completely in detecting the “Achieving” students, as seen from the zero F1-Score for the class.]]>
Copyrights © 2023
