IMPLEMENTASI ORANGE DATA MINING UNTUK KLASIFIKASI KELULUSAN MAHASISWA DENGAN MODEL K-NEAREST NEIGHBOR, DECISION TREE SERTA NAIVE BAYES

  • Hozairi H
  • Anwari A
  • Alim S
N/ACitations
Citations of this article
494Readers
Mendeley users who have this article in their library.

Abstract

Abstrak Proses pemantauan dan evaluasi terhadap kelulusan mahasiswa Universitas Islam Madura (UIM) sangat perlu untuk dilakukan karena tingkat kelulusan mahasiswa merupakan salah satu unsur penilaian akreditasi yang sangat penting untuk setiap Program Studi. Data Mining bisa digunakan untuk klasifikasi ketepatan kelulusan mahasiswa, penelitian ini bertujuan untuk menerapkan aplikasi orange data mining dengan menggunakan model K-Nearest Neighbor (K-NN), Decision Tree serta Naive Bayes dan selanjutnya akan dilakukan evaluasi akurasi dari masing-masing model tersebut. Penelitian ini dilakukan di Prodi Teknik Informatika Universitas Islam Madura tahun angkatan 2016, selanjutnya data mahasiswa akan dianalisa menggunakan aplikasi orange data mining dengan menggunakan model K-NN, Decision Tree serta Naive Bayes. Proses pengujian data menerapkan K-Fold Cross Validation (K=5), sedangkan model evaluasi yang digunakan adalah Confusion Matrix dan ROC. Hasil perbandingan ketiga model sebagai berikut, K-NN memiliki tingkat akurasi sebesar 77%, Decision Tree tingkat akurasi sebesar 74%, dan Naive Bayes memiliki tingkat akurasi sebesar 89%. Maka dari itu, untuk klasifikasi tingkat kelulusan mahasiswa Prodi Teknik Informatika Universitas Islam Madura merekomendasikan model Naive Bayes karena memiliki tingkat akurasi lebih baik dibanding K-NN dan Decision Tree. Abstract The process of monitoring and evaluating the graduation of students at the Islamic University of Madura (UIM) is very necessary because the student's graduation rate is one of the most important elements of accreditation assessment for each Study Program. Data Mining can be used to classify the accuracy of student graduation, this study aims to apply orange data mining applications using the K-Nearest Neighbor (K-NN), Decision Tree and Naive Bayes models and then evaluate the accuracy of each of these models. This research was conducted at the Informatics Engineering Study Program, Islamic University of Madura in class 2016, then student data will be analyzed using an orange data mining application using the K-NN model, Decision Tree and Naive Bayes. The data testing process applies K-Fold Cross Validation (K=5), while the evaluation model used is Confusion Matrix and ROC. The results of the comparison of the three models are as follows, K-NN has an accuracy rate of 77%, Decision Tree has an accuracy rate of 74%, and Naive Bayes has an accuracy rate of 89%. Therefore, for the classification of the graduation rate of students in the Informatics Engineering Study Program, Islamic University of Madura, it is recommended the Naive Bayes model because it has a better accuracy rate than K-NN and Decision Tree.

Cite

CITATION STYLE

APA

Hozairi, H., Anwari, A., & Alim, S. (2021). IMPLEMENTASI ORANGE DATA MINING UNTUK KLASIFIKASI KELULUSAN MAHASISWA DENGAN MODEL K-NEAREST NEIGHBOR, DECISION TREE SERTA NAIVE BAYES. Network Engineering Research Operation, 6(2), 133. https://doi.org/10.21107/nero.v6i2.237

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free