A method for class noise detection based on K-means and SVM algorithms
One of the techniques for improving the accuracy of induced classifier is noise filtering. The classifiers prediction performance is affected by the noisy datasets used in the induction of classifiers. Therefore, it is very important to detect and remove the noise in order to increase the classifica...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/59103/ http://dx.doi.org/10.1007/978-3-319-22689-7_23 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | One of the techniques for improving the accuracy of induced classifier is noise filtering. The classifiers prediction performance is affected by the noisy datasets used in the induction of classifiers. Therefore, it is very important to detect and remove the noise in order to increase the classification accuracy. This paper proposed a model for noise detection in the datasets using k-means and support vector machine (SVM) techniques. The proposed model has been tested using the datasets from University of California, Irvine machine learning repository. Experimental results reveal that the proposed model can improve data quality and increase the classification accuracies. |
|---|