Aboalayon, Khald Ali I. (2005) An Experimental Study of Classification Algorithms Training Performance. Masters thesis, Universiti Utara Malaysia.
Rights:
Restricted
![[thumbnail of KHALD_ALI_I._ABOALAYON.pdf]](https://etd.uum.edu.my/style/images/fileicons/application_pdf.png)
PDF
KHALD_ALI_I._ABOALAYON.pdf
Restricted to Registered users only
Request a copy
KHALD_ALI_I._ABOALAYON.pdf
Restricted to Registered users only
Request a copy
Access Restriction Notice
Full text access is restricted by the author.
Click Request a copy to contact the author by email.
Access is subject to the author's approval.
A hardcopy is available at the Special Collection Counter.
Click Request a copy to contact the author by email.
Access is subject to the author's approval.
A hardcopy is available at the Special Collection Counter.
Preview
Abstract
This thesis evaluates the training performance of classifiers in terms of Root Mean Square Error (RMSE), Training Time and Complexity. The study was based on
different data set that were obtained from UCI machine learning database and tested by the WEKA software machine learning tools. The aim of this study is to experiment several classifiers with different data sets to find out the best classifier for a certain data set like nominal, numerical and both, according to the objective of this research.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisor : | UNSPECIFIED |
| Item ID: | 1250 |
| Uncontrolled Keywords: | Data Mining Techniques, Classification, Classification Algorithms |
| Subjects: | Q Science > QA Mathematics > QA71-90 Instruments and machines |
| Divisions: | Faculty and School System > Faculty of Information Technology |
| Date Deposited: | 02 Feb 2010 03:58 |
| Last Modified: | 24 Jul 2013 12:11 |
| Department: | Department of Computer Science |
| URI: | https://etd.uum.edu.my/id/eprint/1250 |
Actions (login required)
- View Item