<![CDATA[The grouping of student misconduct data for the second semester is used to assess student discipline levels. This data classification uses data mining methods to determine student discipline objectively. The data mining method used in this study is Naïve Bayes. This data classification uses manual calculations with the Gaussian Naïve Bayes method, which uses an integer approach. It is not only tested manually but also with RapidMiner tools. The technique used in Rapid Miner to divide the data into several parts or folds, where the training and testing data parts are divided by cross-validation. This technique aims to make the evaluation results more accurate. The evaluation is made with a confusion matrix with curation, precision, and recall calculations and F1 score. Data grouping is divided into two categories, namely disciplined and undisciplined. The results of the study using Naïve Bayes with GA optimization obtained an accuracy value of 89.47% using the cross-validation technique with stratified sampling type, which helped produce a more stable evaluation.]]>
Copyrights © 2025
