<![CDATA[One factor that influences the quality of college is that student performance can be measured through the length of the study period. The faster study period of student, then better student's performance, and vice versa. Data mining can be interpreted as a process to get useful information from large amount of data so that a knowledge is obtained. This research aims to analyze the comparative performance of probabilistic algorithms namely Naïve Bayes (NB) and Bayesian Network (BN) to classifying the student study period of Universitas Widya Dharma (UNWIDHA) Klaten. Variables for classification are Achievement Index (IP) (1st semester, 2nd semester, and 3rd semester), student school majors and college entrance path. The data used in the form of data are 100 graduates of UNWIDHA. Classification of student study period consisting of study periods <4 years, ≥ 4 & <4.5 years, ≥ 4.5 & <5 years, ≥ 5 & <5.5 years, ≥ 5.5 & <6 and ≥ 6 years. . The results showed that the precision value and recall, BN algorithm were better than NB. On performance measurement with percentage split 90, NB and BN algorithms have the same accuracy value of 80%. Whereas in percentage split 80, BN is superior with an accuracy of 75% while NB is 70%.]]>
Copyrights © 2018
