|Title:||A new algorithm for the reduction of variable precision rough set and a feature weighted SVM with real applications|
|Subject:||Hong Kong Polytechnic University -- Dissertations|
|Department:||Department of Computing|
|Pages:||ix, 89 leaves : ill. ; 30 cm.|
|Abstract:||Rough set theory has been widely applied to deal with different problems since its introduction despite its own limitations. Variable precision rough set is one of the extension theories of classical rough set theory. Unlike the classical rough set theory, the reduction of variable precision rough set theory may lead to the problems of reduction anomalies. What is more, traditional algorithm cannot definitely find the minimal reduction. In this dissertation, a new algorithm which is based on the set covering theory was introduced to solve the reduction problem of variable precision rough set. A feature weighted SVM is also introduced in this dissertation. We give the different weight value to each feature of SVM according to the grey correlation degree value of each feature with respect to the problem. In fact, this method is to change the Euclid distance of each sample point in order to find a better hyper plane. Experiments were conducted to attest that this method can improve the accuracy of classifier. In the final part, we used the stock price data collected from China A share market to do the prediction work of stock price. There are three goals of the experiments: the first is to find a method which combines the rough set theory and SVM theory together which is able to solve the real problem, the second is to attest the improvement of feature weighted SVM in regression problem rather than classification problem, the third is to do research on the generalization capability of the reduction of variable precision rough set.|
|Rights:||All rights reserved|
Files in This Item:
|b24268045.pdf||For All Users (off-campus access for PolyU Staff & Students only)||1.55 MB||Adobe PDF||View/Open|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: