| Autor: |
Wati, Embun Fajar, Sari, Anggi Puspita, Alawiah, Enok Tuti, Siregar, Martua Hami, Rudianto, Biktra |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 5/12/2023, Vol. 2714 Issue 1, p1-11, 11p |
| Abstrakt: |
The pandemic has adversely affected many companies, resulting in many students quitting or taking time off from college because they are no longer working or temporarily laid off. The college must take action in helping students to continue their studies, so that it can be analyzed in the framework of policy making. The purpose of this research is to classify students who are late for graduation, so that the college can provide policies such as scholarships. Data mining classification algorithms used are Decision Tree (C4.5) and Naïve Bayes validated by the 10-fold cross-validation method and optimized by Particle Swarm Optimization in order to improve the performance accuracy of both algorithms. The result of this study is the classification of student graduation data in tfec4 he pandemic can be easily done with Naïve Bayes algorithms tested by 10-fold cross-validation method and optimized with Particle Swarm Optimization method resulting in accuracy of 100.00% superior to Decision Tree algorithm with accuracy of 97.00%. The recall result of both algorithms is the same which is 100%, while the precision with the highest value of 100.00% is achieved by the Naïve Bayes algorithm. Naïve Bayes and Decision Tree also has an excellent AUC score of 1,000. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
