Improved Accuracy by Relearning and Combining Distance Functions

Authors:
Naohiro Ishii;Takahiro Yamada;Yongguang Bao
Affiliations:
Aichi Institute of Technology, Toyota, Japan;Aichi Institute of Technology, Toyota, Japan;Aichi Institute of Technology, Toyota, Japan
Venue:
KES '08 Proceedings of the 12th international conference on Knowledge-Based Intelligent Information and Engineering Systems, Part II
Year:
2008

Citing 4
Cited 0

Text Classification by Combining Different Distance Functions withWeights

SNPD-SAWN '06 Proceedings of the Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing
Data Mining

Data Mining
Improved heterogeneous distance functions

Journal of Artificial Intelligence Research
Classification by instance-based learning algorithm

IDEAL'05 Proceedings of the 6th international conference on Intelligent Data Engineering and Automated Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

The k-nearest neighbor(kNN) is improved by applying the distance functions with relearning and ensemble computations with the higher accuracy values. In this study, the proposed relearning and combining ensemble computations are an effective technique for improving accuracy. We develop a new approach to combine kNN classifier based on different distance functions with relearning and ensemble computations. The proposed combining algorithm shows higher generalization accuracy, compared to our previous studies and other conventional algorithms by artificial intelligence techniques. First, to improve classification accuracy, a relearning method with genetic algorithm is developed. Second, ensemble computations are followed by the relearning. Experiments have been conducted on some benchmark datasets from the UCI Machine Learning Repository.