Welcome to the InfoSci Platform
Reference Hub2
Natural Neighbor Reduction Algorithm for Instance-based Learning

Natural Neighbor Reduction Algorithm for Instance-based Learning

Lijun Yang, Qingsheng Zhu, Jinlong Huang, Dongdong Cheng, Cheng Zhang
Copyright: © 2016 |Volume: 10 |Issue: 4 |Pages: 15
ISSN: 1557-3958|EISSN: 1557-3966|EISBN13: 9781466689671|DOI: 10.4018/IJCINI.2016100103
Cite Article Cite Article

MLA

Yang, Lijun, et al. "Natural Neighbor Reduction Algorithm for Instance-based Learning." IJCINI vol.10, no.4 2016: pp.59-73. http://doi.org/10.4018/IJCINI.2016100103

APA

Yang, L., Zhu, Q., Huang, J., Cheng, D., & Zhang, C. (2016). Natural Neighbor Reduction Algorithm for Instance-based Learning. International Journal of Cognitive Informatics and Natural Intelligence (IJCINI), 10(4), 59-73. http://doi.org/10.4018/IJCINI.2016100103

Chicago

Yang, Lijun, et al. "Natural Neighbor Reduction Algorithm for Instance-based Learning," International Journal of Cognitive Informatics and Natural Intelligence (IJCINI) 10, no.4: 59-73. http://doi.org/10.4018/IJCINI.2016100103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Instance reduction is aimed at reducing prohibitive computational costs and the storage space for instance-based learning. The most frequently used methods include the condensation and edition approaches. Condensation method removes the patterns far from the decision boundary and do not contribute to better classification accuracy, while edition method removes noisy patterns to improve the classification accuracy. In this paper, a new hybrid algorithm called instance reduction algorithm based on natural neighbor and nearest enemy is presented. At first, an edition algorithm is proposed to filter noisy patterns and smooth the class boundaries by using natural neighbor. The main advantage of the algorithm is that it does not require any user-defined parameters. Then, using a new condensation method based on nearest enemy to reduce instances far from decision line. Through this algorithm, interior instances are discarded. Experiments show that the hybrid approach effectively reduces the number of instances while achieves higher classification accuracy along with competitive algorithms.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.