Implementasi Metode Forward Selection pada Algoritma Support Vector Machine (SVM) dan Naive Bayes Classifier Kernel Density (Studi Kasus Klasifikasi Jalur Minat SMA)
| Main Authors: | Sasongko, Theopilus Bayu, Arifin, Oki |
|---|---|
| Format: | Article info Analisis application/pdf eJournal |
| Bahasa: | ind |
| Terbitan: |
Fakultas Ilmu Komputer, Universitas Brawijaya
, 2019
|
| Subjects: | |
| Online Access: |
http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000 http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000/pdf |
| ctrlnum |
article-1000 |
|---|---|
| fullrecord |
<?xml version="1.0"?>
<dc schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd"><title lang="id">Implementasi Metode Forward Selection pada Algoritma Support Vector Machine (SVM) dan Naive Bayes Classifier Kernel Density (Studi Kasus Klasifikasi Jalur Minat SMA)</title><creator lang="id">Sasongko, Theopilus Bayu</creator><creator lang="id">Arifin, Oki</creator><subject lang="id">Classification</subject><subject lang="id">SVM</subject><subject lang="id">Naive Bayes Kernel Density</subject><subject lang="id">Forward Selection</subject><description lang="id">Peminatan merupakan kegiatan yang disediakan oleh pihak sekolah yang berguna untuk mengakomodasi pilihan minat, bakat, atau kemampuan peserta didik dengan orientasi pemusatan. Penentuan jalur minat umumnya melibatkan banyak attribute. Klasifikasi maupun prediksi pada data mining menggunakan fitur seleksi sangat penting untuk pemilihan attribute yang tepat, karena berpengaruh pada performansi model, oleh sebab itu perlu metode untuk melakukan seleksi atribut. Penelitian ini membandingkan implementasi metode forward selection pada Algoritma SVM dan Naïve Bayes Kernel Density. Studi kasus yang digunakan adalah jalur minat pada siswa SMA pada dua sekolah yang berbeda. Proses pembentukan model klasifikasi dengan menganalisa perubahan kernel, faktor pinalti (C) SVM, number of kernel Naïve bayes kernel density, dan hasil feature subset forward selection. Digunakan lima buah eksperimen kernel SVM yaitu dot (linear), radial (RBF), polynomial, neural, dan anova. Proses uji coba perubahan parameter menggunakan rentang 0.0-100.0. Hasil dari penelitian ini diantaranya adalah feature subset dataset SMA ABC yang terpilih yaitu nilai IPA, tes akademik, abstrak konseptual, analisa sintesa, dan logika numerik, sedangkan feature subset SMA XYZ yaitu nilai IPA, logika numerik, dan analisa sintesa. Hasil pengujian dataset SMA ABC menggunakan algoritma FS-SVM berbasis kernel anova parameter C=10.0 sebesar 99.29%. Sedangkan hasil pengujian dataset SMA XYZ menggunakan algoritma FS-SVM berbasis kernel anova parameter C=10.0 sebesar 95.17%. AbstractSpecialization is an activity provided by the school that is useful to accommodate the choice of interests, talents, or abilities of students with a concentration of orientation. The determination of interest generally involved many attributes. The classification and prediction on the data mining that use the selection feature is very important for the selection of the right attribute, because it affects the performance of the model, therefore a method is needed to select attributes. This study compares the implementation of the forward selection method in the SVM Algorithm and Naïve Bayes Kernel Density. The case study that is used is the interest of students in high school and compared with two different schools. The process of modelling by studying kernel changes, penalty factors (C) SVM, number of kernel Naïve bayes kernels, and the results of features from subset forward selection. Five SVM kernel experiments ared used, namely dot (linear), radial (RBF), polynomial, neural, and anova. The trial process of changes parameters uses the range 0.0-100.0. The results of this study include features of selected ABC SMA subset datas, which are IPA values, academic tests, conceptual abstracts, synthesis analysis, and numerical logic, while the XYZ SMA subset features are IPA values, numerical logic, and synthesis analysis. The test results of the ABC High School dataset that use the kernel-based FS-SVM algorithm parameter C = 10.0 is 99.29%. While the results of testing the XYZ SMA dataset that use the kernel-based FS-SVM algorithm parameter C = 10.0 for 95.17%.</description><publisher lang="en">Fakultas Ilmu Komputer, Universitas Brawijaya</publisher><date>2019-07-15</date><type>Journal:Article</type><type>Other:info:eu-repo/semantics/publishedVersion</type><type>Other:Analisis</type><type>File:application/pdf</type><identifier>http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000</identifier><source lang="id">Jurnal Teknologi Informasi dan Ilmu Komputer; Vol 6 No 4: Agustus 2019; 383-388</source><source lang="en">Jurnal Teknologi Informasi dan Ilmu Komputer; Vol 6 No 4: Agustus 2019; 383-388</source><source>2528-6579</source><source>2355-7699</source><source>10.25126/jtiik.201964</source><language>ind</language><relation>http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000/pdf</relation><rights lang="en">Hak Cipta (c) 2019 Jurnal Teknologi Informasi dan Ilmu Komputer</rights><recordID>article-1000</recordID></dc>
|
| language |
ind |
| format |
Journal:Article Journal Other:info:eu-repo/semantics/publishedVersion Other Other:Analisis File:application/pdf File Journal:eJournal |
| author |
Sasongko, Theopilus Bayu Arifin, Oki |
| title |
Implementasi Metode Forward Selection pada Algoritma Support Vector Machine (SVM) dan Naive Bayes Classifier Kernel Density (Studi Kasus Klasifikasi Jalur Minat SMA) |
| publisher |
Fakultas Ilmu Komputer, Universitas Brawijaya |
| publishDate |
2019 |
| topic |
Classification SVM Naive Bayes Kernel Density Forward Selection |
| url |
http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000 http://jtiik.ub.ac.id/index.php/jtiik/article/view/1000/pdf |
| contents |
Peminatan merupakan kegiatan yang disediakan oleh pihak sekolah yang berguna untuk mengakomodasi pilihan minat, bakat, atau kemampuan peserta didik dengan orientasi pemusatan. Penentuan jalur minat umumnya melibatkan banyak attribute. Klasifikasi maupun prediksi pada data mining menggunakan fitur seleksi sangat penting untuk pemilihan attribute yang tepat, karena berpengaruh pada performansi model, oleh sebab itu perlu metode untuk melakukan seleksi atribut. Penelitian ini membandingkan implementasi metode forward selection pada Algoritma SVM dan Naïve Bayes Kernel Density. Studi kasus yang digunakan adalah jalur minat pada siswa SMA pada dua sekolah yang berbeda. Proses pembentukan model klasifikasi dengan menganalisa perubahan kernel, faktor pinalti (C) SVM, number of kernel Naïve bayes kernel density, dan hasil feature subset forward selection. Digunakan lima buah eksperimen kernel SVM yaitu dot (linear), radial (RBF), polynomial, neural, dan anova. Proses uji coba perubahan parameter menggunakan rentang 0.0-100.0. Hasil dari penelitian ini diantaranya adalah feature subset dataset SMA ABC yang terpilih yaitu nilai IPA, tes akademik, abstrak konseptual, analisa sintesa, dan logika numerik, sedangkan feature subset SMA XYZ yaitu nilai IPA, logika numerik, dan analisa sintesa. Hasil pengujian dataset SMA ABC menggunakan algoritma FS-SVM berbasis kernel anova parameter C=10.0 sebesar 99.29%. Sedangkan hasil pengujian dataset SMA XYZ menggunakan algoritma FS-SVM berbasis kernel anova parameter C=10.0 sebesar 95.17%. AbstractSpecialization is an activity provided by the school that is useful to accommodate the choice of interests, talents, or abilities of students with a concentration of orientation. The determination of interest generally involved many attributes. The classification and prediction on the data mining that use the selection feature is very important for the selection of the right attribute, because it affects the performance of the model, therefore a method is needed to select attributes. This study compares the implementation of the forward selection method in the SVM Algorithm and Naïve Bayes Kernel Density. The case study that is used is the interest of students in high school and compared with two different schools. The process of modelling by studying kernel changes, penalty factors (C) SVM, number of kernel Naïve bayes kernels, and the results of features from subset forward selection. Five SVM kernel experiments ared used, namely dot (linear), radial (RBF), polynomial, neural, and anova. The trial process of changes parameters uses the range 0.0-100.0. The results of this study include features of selected ABC SMA subset datas, which are IPA values, academic tests, conceptual abstracts, synthesis analysis, and numerical logic, while the XYZ SMA subset features are IPA values, numerical logic, and synthesis analysis. The test results of the ABC High School dataset that use the kernel-based FS-SVM algorithm parameter C = 10.0 is 99.29%. While the results of testing the XYZ SMA dataset that use the kernel-based FS-SVM algorithm parameter C = 10.0 for 95.17%. |
| id |
IOS577.article-1000 |
| institution |
Universitas Brawijaya |
| institution_id |
30 |
| institution_type |
library:university library |
| library |
Perpustakaan Universitas Brawijaya |
| library_id |
480 |
| collection |
Jurnal Teknologi Informasi dan Ilmu Komputer |
| repository_id |
577 |
| subject_area |
Program Komputer dan Teknologi Informasi |
| city |
MALANG |
| province |
JAWA TIMUR |
| repoId |
IOS577 |
| first_indexed |
2019-07-20T01:36:56Z |
| last_indexed |
2024-06-02T20:42:41Z |
| recordtype |
dc |
| merged_child_boolean |
1 |
| _version_ |
1800783725749862400 |
| score |
17.13294 |