Peramalan Dengan Menggunakan Artificial Neural Network ( Ann ) Dan Support Vector Regression ( Svr )

Nugraha, Riza (2007) Peramalan Dengan Menggunakan Artificial Neural Network ( Ann ) Dan Support Vector Regression ( Svr ). Other thesis, Institut Teknologi Sepuluh Nopember Surabaya.

Text 2502109044-Undergraduate_Thesis.pdf - Accepted Version Download (12MB)

Abstract

Support Vector Regression merupakan teknik yang relative baru untuk peramalan. Dalam paper ini Support Vector Regression diuji untuk beberapa set data baik data time series atau non time series. Sebagai pembanding untuk menilai performansi Support Vector Regression, diimplementasikan juga Artificial Neural Network (ANN), Double explonential smoothing dan trend analysis. Set data yang digunakan adalah harga saham, harga rumah , data konsumsi bahan bakar kendaraan , dan harga mobil . Pengujian dilakukan dengan bantuan software Minitab 14 dan Matlab 7.0. Penelitian ini menghasilkan kesimpulan bahwa tidak ada metode terbaik dalam melakukan peramalan tetapi ukuran data training yang digunakan akan mempengaruhi nilai akurasi dari peramalan tersebut. Parameter yang digunakan untuk penilaian hasil uji adalah nilai mean square error . Pada umumnya Artificial neural networks memberi hasil yang lebih baik dibanding metoda yang lain
=====================================================================================================================================
Support Vector Regression is a relatively new technique for forecasting. In this paper Support Vector Regression is tested for several data sets, both time series and non-time series data. As a comparison to assess the performance of Support Vector Regression, it is also implemented Artificial Neural Network (ANN), Double explonential smoothing and trend analysis. The data sets used are stock prices, house prices, vehicle fuel consumption data, and car prices. Testing was carried out with the help of Minitab 14 conclusion that there is no best method for forecasting but the size of the training data used will affect the accuracy of the forecast. The parameter used for the assessment of test results is the mean square error value. In general, Artificial neural networks give better results than other methods

Item Type:	Thesis (Other)
Additional Information:	RSI 658.403 55 Nug p
Uncontrolled Keywords:	Artificial Neural Network (ANN), Support Vector Regression (SVR), mean square error (mse)
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > T Technology (General) > T174 Technological forecasting
Divisions:	Faculty of Industrial Technology > Industrial Engineering > 26201-(S1) Undergraduate Thesis
Depositing User:	EKO BUDI RAHARJO
Date Deposited:	11 Jan 2023 06:13
Last Modified:	11 Jan 2023 06:13
URI:	http://repository.its.ac.id/id/eprint/95367

Actions (login required)

View Item