FuSCa: A New Weighted Membership Driven Fuzzy Supervised Classifier
The aim of this paper is to introduce a new supervised fuzzy classification methodology (FuSCa) to improve the performance of k-NN (k-Nearest Neighbor) algorithm based on the weighted nearest neighbor membership and global membership derived from the training dataset. In this classification method, the test object is assigned a class label having the maximum membership value for that corresponding class while a weighted membership vector is found after utilizing the Global and Nearest-Neighbor fuzzy membership vectors along with a global weight and a k-close weight respectively. FuSCa is compared with other approaches using the standard benchmark data-sets and found to produce better classification accuracy.
KeywordsFuzzy membership Supervised classification Machine learning Data-mining Nearest neighbor Weighted membership
Unable to display preview. Download preview PDF.
- 5.Chung-Hoon, F., Hwang, C.: An interval type-2 fuzzy k-nearest neighbor. In: Proceedings of the 12th IEEE International Conference on Fuzzy Systems, pp. 802–807 (2003)Google Scholar
- 11.Tan, P.-N., Steinbach, M., Kumar, V.: Introduction to Data Mining. Pearson Education, Inc. (2006)Google Scholar
- 13.Witten, H.: Data Mining: Practical Machine Learning Tools and Techniques. Morgan Kaufmann Series in Data Management Systems (2005)Google Scholar
- 14.WEKA software, Machine Learning, The University of Waikato, Hamilton, New Zealand, http://www.cs.waikato.ac.nz/ml/weka/
- 15.UCI Machine Learning Repository, http://mlearn.ics.uci.edu/MLRepository.html