Abstract
Distance classifier ensemble method based on Intra-class and Inter-class Scatter is proposed in this paper. By Bootstrap technology, the training samples are sampled repeatedly to generate several subsample set, define Intra-class and Inter-class Scatter matrix with subsample set, train subsample set with scatter matrix, generate individual classifier. In the classifier ensemble, the results are integrated with the relative majority voting method. Experiment is tested on UCI standard database, the experimental results show that the proposed ensemble method based on Intra-class and Inter-class Scatter for distance classifier is effective, and it is superior to other methods in classification performance.
Highlights
In the field of pattern recognition, years of practical experience shows that, it is difficult to obtain satisfactory recognition performance by a the single method for complex pattern recognition problems[1,2,3]
The results are integrated with the relative majority voting method
Experiment is tested on UCI standard database, the paper method is compared with minimum distance classifier integration method
Summary
In the field of pattern recognition, years of practical experience shows that, it is difficult to obtain satisfactory recognition performance by a the single method for complex pattern recognition problems[1,2,3]. Using Bagging technology, individual classifier training samples are selected from the original training set. By reselecting the training set, the Bagging technique increases classifier integration difference degree, and improve the generalization ability. Distance classifier ensemble method is proposed in this paper. The training samples can be sampled repeatedly to generate several different subsample sets, the subsample set is used to establish the Intra-class and Inter-class scatter matrix, train subsample set with scatter matrix, generate individual classifier. Experiment is tested on UCI standard database, the paper method is compared with minimum distance classifier integration method. The performance of single minimum distance classifier and single minimum distance classifier based on Intra-class and Inter-class scatter are compared with the proposed method, the experimental results show that the proposed ensemble method is effective
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: DEStech Transactions on Computer Science and Engineering
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
