ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK
There are many algorithm that can be used to predict traffic flow but it is not known which algorithm that has a more accurate performance. So that each algorithm needs to be tested to find out. The proposed method is an accuracy comparison method of algorithm-based neural network that can be used f...
Main Author: | |
---|---|
Format: | Article |
Language: | Indonesian |
Published: |
LPPM Universitas Potensi Utama
2014-10-01
|
Series: | CSRID Journal |
Subjects: | |
Online Access: | http://csrid.potensi-utama.ac.id/ojs/index.php/CSRID/article/view/28/pdf |
id |
doaj-b672e44844e04a83be5c4f94ff8b1204 |
---|---|
record_format |
Article |
spelling |
doaj-b672e44844e04a83be5c4f94ff8b12042020-11-24T20:59:19ZindLPPM Universitas Potensi UtamaCSRID Journal2085-13672460-870X2014-10-0163148158ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEKBambang Lareno0Teknik Informatika, STMIK Indonesia Jl. P. Hidayatullah (Samping Jembatan Banua Anyar) Banjarmasin Telp.(0511) 4315530There are many algorithm that can be used to predict traffic flow but it is not known which algorithm that has a more accurate performance. So that each algorithm needs to be tested to find out. The proposed method is an accuracy comparison method of algorithm-based neural network that can be used for predictive a time-series data. Algorithms to be tested is the back Propagation Neural Network (BP-NN), Adaptive Neuro Fuzzy Inference System (ANFIS), Wavelet Neural Network (WNN), dan Evolving Neural Network (ENN), which is used to predict short-term traffic flow. Each algorithm will be implemented using MatLab 2009b. Performance measurement is done by calculating the average amount of error that occurred through the Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Absolute Deviation (MAD). The smaller value of each performance parameter will indicate that the predictive value was closer to the true value. Thus a more accurate algorithm can be determined. The test results determine that ENN is more accurate prediction algorithms for short-term traffic flow.http://csrid.potensi-utama.ac.id/ojs/index.php/CSRID/article/view/28/pdfTraffic Flow PredictionTime-Series Forecasting |
collection |
DOAJ |
language |
Indonesian |
format |
Article |
sources |
DOAJ |
author |
Bambang Lareno |
spellingShingle |
Bambang Lareno ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK CSRID Journal Traffic Flow Prediction Time-Series Forecasting |
author_facet |
Bambang Lareno |
author_sort |
Bambang Lareno |
title |
ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK |
title_short |
ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK |
title_full |
ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK |
title_fullStr |
ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK |
title_full_unstemmed |
ANALISA DAN PERBANDINGAN AKURASI MODEL PREDIKSI RENTET WAKTU ARUS LALU LINTAS JANGKA PENDEK |
title_sort |
analisa dan perbandingan akurasi model prediksi rentet waktu arus lalu lintas jangka pendek |
publisher |
LPPM Universitas Potensi Utama |
series |
CSRID Journal |
issn |
2085-1367 2460-870X |
publishDate |
2014-10-01 |
description |
There are many algorithm that can be used to predict traffic flow but it is not known which algorithm that has a more accurate performance. So that each algorithm needs to be tested to find out. The proposed method is an accuracy comparison method of algorithm-based neural network that can be used for predictive a time-series data. Algorithms to be tested is the back Propagation Neural Network (BP-NN), Adaptive Neuro Fuzzy Inference System (ANFIS), Wavelet Neural Network (WNN), dan Evolving Neural Network (ENN), which is used to predict short-term traffic flow. Each algorithm will be implemented using MatLab 2009b. Performance measurement is done by calculating the average amount of error that occurred through the Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Absolute Deviation (MAD). The smaller value of each performance parameter will indicate that the predictive value was closer to the true value. Thus a more accurate algorithm can be determined. The test results determine that ENN is more accurate prediction algorithms for short-term traffic flow. |
topic |
Traffic Flow Prediction Time-Series Forecasting |
url |
http://csrid.potensi-utama.ac.id/ojs/index.php/CSRID/article/view/28/pdf |
work_keys_str_mv |
AT bambanglareno analisadanperbandinganakurasimodelprediksirentetwaktuaruslalulintasjangkapendek |
_version_ |
1716782839935008768 |