원문정보
초록
영어
This paper presents a new hybrid classifier that combines the k-Nearest Neighbor (k-NN) distance based algorithm with the classification tree paradigm based on the ID3 algorithm. The k-NN algorithm is used as a preprocessing algorithm in order to obtain a modified training database for the posterior learning of the classification tree structure. Then the incorrectly classified instances are duplicated with the previous data set and finally ID3 is applied to complete the classification procedure of biomedical data. In this approach a boosting technique is incorporated in such way that the incorrectly classified instances in the training set are identified using the k –NN algorithm. The performance of the proposed method is compared with the related algorithms. Experimental results show that the newly proposed approach performs better than the other existing techniques.
목차
1. Introduction
2. Overview of the Work
3. Decision Tree Algorithm
3.1. Decision Tree Induction Using ID3 Algorithm
3.2. Attribute Selection
3.3. Tree Pruning
3.4. Review Stage
3.5. Boosting Using k-NN Algorithm
4. Features of k-NN
4.1. Decision Rule and Confusion Matrix for Classification
4.2. Performance Assessment with Cross-Validation
5. Result and Discussion
6. Conclusions
References