View Item 
  •   Unmul Repository Home
  • Akuntabel
  • J - Computer Sciences and Information Technology
  • View Item
  •   Unmul Repository Home
  • Akuntabel
  • J - Computer Sciences and Information Technology
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Time Series Prediction Using Radial Basis Function Neural Network

Thumbnail
View/Open
15. 5671-4891-1-PB.pdf (424.2Kb)
Date
2015-08-05
Author
Haviluddin, Haviluddin
Tahyudin, Imam
Metadata
Show full item record
Abstract
This paper presents an approach for predicting daily network traffic using artificial neural networks (ANN), namely radial basis function neural network (RBFNN) method. The data is gained from 21-24 June 2013 (192 samples series data) in ICT Unit of Mulawarman University, East Kalimantan, Indonesia. The results of measurement are using statistical analysis, e.g. sum of square error (SSE), mean of square error (MSE), mean of absolute percentage error (MAPE), and mean of absolute deviation (MAD). The results show that values are the same, with different goals that have been set are 0.001, 0.002, and 0.003, and spread 200. The smallest MSE value indicates a good method for accuracy. Therefore, the RBFNN model illustrates the proposed best model to predict daily network traffic.
URI
http://repository.unmul.ac.id/handle/123456789/3262
Collections
  • J - Computer Sciences and Information Technology [94]

Repository Universitas Mulawarman copyright ©   LP3M Universitas Mulawarman
Jalan Kuaro Kotak Pos 1068
Telp. (0541) 741118
Fax. (0541) 747479 - 732870
Samarinda 75119, Kalimantan Timur, Indonesia
Contact Us | Send Feedback
 

 

Browse

All of Unmul RepositoryCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

My Account

Login

Repository Universitas Mulawarman copyright ©   LP3M Universitas Mulawarman
Jalan Kuaro Kotak Pos 1068
Telp. (0541) 741118
Fax. (0541) 747479 - 732870
Samarinda 75119, Kalimantan Timur, Indonesia
Contact Us | Send Feedback