Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476385Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Salwani Abdullah, Prof. Dr | |
| dc.contributor.author | Ali Hassan Ali Abdullah (P65677) | |
| dc.date.accessioned | 2023-10-06T09:17:31Z | - |
| dc.date.available | 2023-10-06T09:17:31Z | - |
| dc.date.issued | 2013-12-20 | |
| dc.identifier.other | ukmvital:84846 | |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/476385 | - |
| dc.description | Time series prediction is an active research area that has drawn considerable attention during the last decades for a variety of applications. A time series is a collection of data recorded over a period of time such as on weekly, monthly, quarterly, or yearly basis. Researchers have introduced models such as neural network, genetic algorithm, support vector machine that aim to achieve high prediction accuracy and efficiency. However, researches indicated that most of models suffer from a number of shortcomings such as easily trapped into a local optimum, premature convergence, and high computation complexity. In order to tackle the listed shortcomings, the research presented in this thesis provides a hybrid approach in obtaining better prediction accuracies. The research firstly aims to investigate a new times series prediction method that is based on Radial Basis Function (RBF) and a Particle Swarm Optimization (PSO) algorithm (coded as RBF-PSO). The approach is applied on Mackey-Glass Time Series (MGTS) and Competition on Artificial Time Series (CATS) datasets to choose a most suitable RBF structure design that can produce an efficient outcome. The obtained results are not comparable with state-of-the-arts. Thus, the second aim is to enhance the predicting ability of the algorithm by employing a Taguchi method together with a Minitab to tune and optimize the parameters for RBF-PSO. To further verify the enhanced approach, a real world dataset called the Rainfall dataset is used. The performance of the proposed method is measured by calculating the Means Square Error (MSE) and Normalized Rooted Means Square Error (NRMSE). The results revealed that the RBF-PSO-TM yields competitive results in comparison with other methods tested on the same datasets, if not the best for MGTS case. The results also demonstrate that the proposed method is able to produce good prediction accuracy when tested on real world rainfall dataset as well.,Master / Sarjana | |
| dc.language.iso | eng | |
| dc.publisher | UKM, Bangi | |
| dc.relation | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
| dc.rights | UKM | |
| dc.subject | Time series prediction | |
| dc.subject | Heuristic algorithms. | |
| dc.title | Hybridization between radial basis function and particle swarm optimisation for time series prediction problems | |
| dc.type | theses | |
| dc.format.pages | 88 | |
| dc.identifier.callno | QA280.A435 2013 | |
| dc.identifier.barcode | 002005 | |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_84846+SOURCE1+SOURCE1.0.PDF Restricted Access | 2.4 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.