Garuda - Garba Rujukan Digital

Hasanuddin -

Unknown Affiliation

Author-ID : 207338

Agriculture, Biological Sciences & Forestry Civil Engineering, Building, Construction & Architecture Earth & Planetary Sciences Economics, Econometrics & Finance Education Environmental Science Law, Crime, Criminology & Criminal Justice Social Sciences Other

Published : 7 Documents Claim Missing Document

Claim Missing Document

Articles

Title

1

PEMBELAJARAN JARINGAN SYARAF TIRUAN PROBABILISTIK BASIS RADIAL DENGAN MENGGUNAKAN ANALISIS SENSITIVITAS -, Hasanuddin
Gamatika Vol 3, No 2 (2012): Jurnal Gagasan Matematika Dan Informatika
Publisher : Gamatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (293.291 KB)

Abstrak Makalah ini menyajikan algoritma pembelajaran bagi Radial Basis Probabilistic Neural Network berdasarkan keturunan gradien fungsi kesalahan. RBPNN terintegrasi dari jaringan saraf radial fungsi dasar (RBFNN) dan jaringan syaraf probabilistik (PNN). Langkah Kepekaan terhadap masukan atas dataset pelatihan berdasarkan turunan parsial dari model RBPNN dan aturan untuk memilih fitur berlebihan juga hadir. Analisis sensitivitas metode yang dapat meningkatkan efisiensi dan efektivitas jaringan saraf. Akhirnya, untuk mengevaluasi kinerja, pendekatan yang diusulkan kami menunjukkan melalui memberikan dua contoh kehidupan nyata kumpulan data. Kata Kunci: RBPNN, keturunan gradien, analisis sensitivitas, seleksi fitur, klasifikasi. Abstract This paper presents a learning algorithm for Radial Basis Probabilistic Neural Network based on gradient descent of error functions. RBPNN integrates of radial basis function neural networks (RBFNN) and probabilistic neural networks (PNN). Sensitivity measures to input over training dataset based on partial derivative of the RBPNN model and rule for selecting redundant feature is also present. Sensitivity analysis of that method can improved efficiency and effectiveness of the neural networks. Finally, to evaluate the performance, our proposed approaches are demonstrated trough giving two examples of real life data set. Keyword: RBPNN, gradient descent, sensitivity analysis, feature selection, classification.

PEMBELAJARAN JARINGAN SYARAF TIRUAN PROBABILISTIK BASIS RADIAL DENGAN MENGGUNAKAN ANALISIS SENSITIVITAS -, Hasanuddin
Gamatika Vol 3, No 2 (2013): Jurnal Gagasan Matematika Dan Informatika
Publisher : Gamatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (293.291 KB)

Abstrak Makalah ini menyajikan algoritma pembelajaran bagi Radial Basis Probabilistic Neural Network berdasarkan keturunan gradien fungsi kesalahan. RBPNN terintegrasi dari jaringan saraf radial fungsi dasar (RBFNN) dan jaringan syaraf probabilistik (PNN). Langkah Kepekaan terhadap masukan atas dataset pelatihan berdasarkan turunan parsial dari model RBPNN dan aturan untuk memilih fitur berlebihan juga hadir. Analisis sensitivitas metode yang dapat meningkatkan efisiensi dan efektivitas jaringan saraf. Akhirnya, untuk mengevaluasi kinerja, pendekatan yang diusulkan kami menunjukkan melalui memberikan dua contoh kehidupan nyata kumpulan data. Kata Kunci: RBPNN, keturunan gradien, analisis sensitivitas, seleksi fitur, klasifikasi. Abstract This paper presents a learning algorithm for Radial Basis Probabilistic Neural Network based on gradient descent of error functions. RBPNN integrates of radial basis function neural networks (RBFNN) and probabilistic neural networks (PNN). Sensitivity measures to input over training dataset based on partial derivative of the RBPNN model and rule for selecting redundant feature is also present. Sensitivity analysis of that method can improved efficiency and effectiveness of the neural networks. Finally, to evaluate the performance, our proposed approaches are demonstrated trough giving two examples of real life data set. Keyword: RBPNN, gradient descent, sensitivity analysis, feature selection, classification.

PEMBELAJARAN JARINGAN SYARAF TIRUAN PROBABILISTIK BASIS RADIAL DENGAN MENGGUNAKAN ANALISIS SENSITIVITAS -, Hasanuddin
Gamatika Vol 3, No 2 (2013): Jurnal Gagasan Matematika Dan Informatika
Publisher : Gamatika

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstrak Makalah ini menyajikan algoritma pembelajaran bagi Radial Basis Probabilistic Neural Network berdasarkan keturunan gradien fungsi kesalahan. RBPNN terintegrasi dari jaringan saraf radial fungsi dasar (RBFNN) dan jaringan syaraf probabilistik (PNN). Langkah Kepekaan terhadap masukan atas dataset pelatihan berdasarkan turunan parsial dari model RBPNN dan aturan untuk memilih fitur berlebihan juga hadir. Analisis sensitivitas metode yang dapat meningkatkan efisiensi dan efektivitas jaringan saraf. Akhirnya, untuk mengevaluasi kinerja, pendekatan yang diusulkan kami menunjukkan melalui memberikan dua contoh kehidupan nyata kumpulan data. Kata Kunci: RBPNN, keturunan gradien, analisis sensitivitas, seleksi fitur, klasifikasi. Abstract This paper presents a learning algorithm for Radial Basis Probabilistic Neural Network based on gradient descent of error functions. RBPNN integrates of radial basis function neural networks (RBFNN) and probabilistic neural networks (PNN). Sensitivity measures to input over training dataset based on partial derivative of the RBPNN model and rule for selecting redundant feature is also present. Sensitivity analysis of that method can improved efficiency and effectiveness of the neural networks. Finally, to evaluate the performance, our proposed approaches are demonstrated trough giving two examples of real life data set. Keyword: RBPNN, gradient descent, sensitivity analysis, feature selection, classification.

Title Search

Found 3 Documents Search Journal : Gamatika

Abstract

Abstract

Abstract

Title

Found 3 Documents
Search
Journal : Gamatika