Knowledge Discovery from Granule Features Mining

Jian-hong Luo (), Xi-yong Zhu and Xiao-jun Wang
Additional contact information
Jian-hong Luo: Zhejiang Sci-Tech University
Xi-yong Zhu: Zhejiang Sci-Tech University
Xiao-jun Wang: Zhejiang Sci-Tech University

Chapter Chapter 40 in The 19th International Conference on Industrial Engineering and Engineering Management, 2013, pp 391-401 from Springer

Abstract: Abstract For the learning problem on imbalanced distribution of data sets, traditional machine learning algorithms tend to produce poor predictive accuracy over the minority class. In this paper, granule features mining model (GFMM) for knowledge discovery is proposed to improve classification accuracy on the minority class. Suitable information granules (IGs) are constructed by ETM-ART2, and then key features analysis method is proposed to discrete represent the IGs to mining compact knowledge rules. The final class for new samples inputted to GFMM can soon be decided by the knowledge rules. Experiments were conducted on three data sets with different skewed level, the results show that GFMM can lead to significant improvement on classification performance for the minority class and outperforms individual SVM and C4.5.

Keywords: Classification; Data mining; Information granules; Key features analysis (search for similar items in EconPapers)
Date: 2013
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-642-38391-5_40

Ordering information: This item can be ordered from
http://www.springer.com/9783642383915

DOI: 10.1007/978-3-642-38391-5_40

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().