Klasifikasi Data Dengan Jaringan Syaraf Fungsi Basis Radial

Budhiarti, Vasthy (2010) Klasifikasi Data Dengan Jaringan Syaraf Fungsi Basis Radial. Masters thesis, Institut Teknologi Sepuluh November.

[thumbnail of 1208201008-Master_thesis.pdf] Text
1208201008-Master_thesis.pdf
Restricted to Repository staff only

Download (11MB)

Abstract

Jaringan Syaraf Fungsi Basis Radial (RBFNN: Radial Basis Function Neural Network) dikenal sebagai salah satu bentuk dari Jaringan Syaraf Feedforward Lapis Banyak (MFNN: Multilayer Feedforward Neural Network), yang handal dalam memecahkan masalah aproksimasi fungsi atau regresi dan klasifikasi data. Tesis ini membahas tentang teknik klasifikasi data dengan arsitektur RBFNN dan diuji dalam proses penjurusan siswa SMA. Penelitian ini diawali dengan pengambilan data sampel berupa nilai rapor siswa kelas X dan beberapa data pendukung semester genap untuk TA 2005/2006 sampai dengan TA 2008/2009 di SMAN 3 Surabaya. Nilai rapor dan data angket siswa diambil sebagai data input karena kedua data tersebut dianggap sudah mewakili bakat, minat dan kemampuan dari siswa. Dengan data target dalam pelatihan ini diperoleh dari data hasil penjurusan yang dilakukan oleh yang berkompeten (guru) yaitu data siswa yang telah kelas XI dan diklasifikasikan dalam jurusan (IPA, IPS dan BAHASA). Langkah selanjutnya, data diproses sehingga dapat diterapkan untuk pelatihan dan pengujian RBFNN. Di mana sebagai pembanding juga digunakan pelatihan dan pengujian MFNN lainnya, yaitu Jaringan Backpropagation (BP). Di mana kedua pelatihan dan pengujian tersebut diterapkan dalam bentuk kode program MATLAB. Hasil dari kedua pelatihan dan pengujian tersebut adalah tingkat akurasi hasil klasifikasi data. Jadi, hasil yang diperoleh dalam penelitian ini yaitu hasil dari klasifikasi data dengan RBFNN memiliki tingkat akurasi yang cukup tinggi dibandingkan hasil klasifikasi data dengan BP. Oleh karena itu, klasifikasi data dengan RBFNN dapat digunakan sebagai alternatif untuk membantu sekolah dalam rangka proses penjurusan siswa SMA.
====================================================================================================================================
Radial Basis Function Neural Network (RBFNN) is known as one form of Multilayer Feedforward Neural Network (MFNN), which is reliable in solving function approximation or regression problems and data classification. This thesis studies the technique of data classification with RBFNN architecture and is tested in the process of senior high school students' majors. This research was initiated with a sample data collection in the form of report card scores of tenth-grade students and some supporting data for the second semester of the academic year 2005/2006 to 2008/2009 at SMAN 3 Surabaya. Report card scores and student questionnaire data were taken as input data because both data were considered to represent the talents, interests, and abilities of the students. The target data in the training were obtained from the major selection data conducted by the competent party (teachers), namely data of students who had reached grade XI and were classified into majors (Science, Social Studies, and Language). The next step, the data were processed so that they could be applied to RBFNN training and testing. As a comparison, training and testing of another MFNN, namely the Backpropagation Network (BP), were also applied. Both training and testing were implemented in the form of MATLAB code. The results from both training and testing are the accuracy levels of each architecture in data classification. Thus, the results obtained in this study show that data classification with RBFNN has a fairly high degree of accuracy compared with data classification using BP. Therefore, data classification with RBFNN can be used as an alternative to assist schools in the process of selecting senior high school students' majors.

Item Type: Thesis (Masters)
Additional Information: 006.312 Bud k
Uncontrolled Keywords: Jaringan Syaraf Tiruan, Klasifikasi, MFNN, RBFNN, BP, Proses Penjurusan siswa SMA, Neural Network, Classification, MFNN, RBFNN, BP, Process majors of senior high school students.
Subjects: Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions: Faculty of Mathematics and Science > Mathematics > 44101-(S2) Master Thesis
Depositing User: magang .
Date Deposited: 30 Sep 2026 07:38
Last Modified: 30 Sep 2026 07:38
URI: http://repository.its.ac.id/id/eprint/145098

Actions (login required)

View Item View Item