The Addition of Adaboost to The Use of The C4.5 Algorithm to Improve The Accuracy of Classification of Study Interests

Specialization is one of the important things in focusing the student's field of study. At one university that has a faculty of computer science, there is an Informatics Engineering undergraduate study program which is divided into two specializations, namely intelligent systems and software en...

Full description

Saved in:
Bibliographic Details
Published in:Indonesian Journal of Information Systems 2024-02, Vol.6 (2), p.130-139
Main Authors: Fahriah, Sirli, Nur Diyana Kamarudin, Liliek Triyono, Adhy Rizaldy
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Specialization is one of the important things in focusing the student's field of study. At one university that has a faculty of computer science, there is an Informatics Engineering undergraduate study program which is divided into two specializations, namely intelligent systems and software engineering design. At one university that has a faculty of computer science, there is a bachelor's program in Informatics Engineering which is divided into two specializations, namely intelligent systems and software engineering design. students find it difficult to choose one of the specializations in the informatics engineering study program. To overcome these problems, the authors provide solutions in the form of ideas that can recommend students in determining specialization. In this problem, the algorithm that will be used is the C4.5 algorithm based on forward selection plus adaboost. The results of the specialization classification use the selected attributes and iterate over the cross-validation so as to produce the right accuracy. Testing the C4.5 algorithm produces an accuracy of 93.89% and the C4.5 algorithm based on forward selection produces an accuracy of 94.44% while using the C4.5 algorithm based on forward selection with the addition of adaboost produces an accuracy of 94.63%. Based on these tests, it proves that there is an increase in accuracy by adding selection and adaboost features to the C4.5 algorithm.
ISSN:2623-0119
2623-2308
DOI:10.24002/ijis.v6i2.7588