Projection with Double Nonlinear Integrals for Classification

Authors:
Jinfeng Wang;Kwongsak Leung;Kinhong Lee;Zhenyuan Wang
Affiliations:
Department of Computer Science & Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong SAR;Department of Computer Science & Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong SAR;Department of Computer Science & Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong SAR;Department of Mathematics, University of Nebraska at Omaha, Omaha, USA NE 68182
Venue:
ICDM '08 Proceedings of the 8th industrial conference on Advances in Data Mining: Medical Applications, E-Commerce, Marketing, and Theoretical Aspects
Year:
2008

Citing 8
Cited 0

Non-monotonic fuzzy measures and the Choquet integral

Fuzzy Sets and Systems
Classification by fuzzy integral: performance and tests

Fuzzy Sets and Systems - Special issue on fuzzy methods for computer vision and pattern recognition
The representation of importance and interaction of features by fuzzy measures

Pattern Recognition Letters - Special issue on fuzzy set technology in pattern recognition
A genetic algorithm for determining nonadditive set functions in information fusion

Fuzzy Sets and Systems - Special issue on fuzzy measures and integrals
Fuzzy Measure Theory

Fuzzy Measure Theory
Universal Artificial Intelligence: Sequential Decisions Based On Algorithmic Probability

Universal Artificial Intelligence: Sequential Decisions Based On Algorithmic Probability
Genetic algorithms for determining fuzzy measures from data

Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology
Classification by nonlinear integral projections

IEEE Transactions on Fuzzy Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this study, a new classification model based on projection with Double Nonlinear Integrals is proposed. There exist interactions among predictive attributes towards the decisive attribute. The contribution rate of each combination of predictive attributes, including each singleton, towards the decisive attribute can be re presented by a fuzzy measure. We use Double Nonlinear Integrals with respect to the signed fuzzy measure to project data to 2-Dimension space. Then classify the virtual value in the 2-D space projected by Nonlinear Integrals. In our experiments, we compare our classifier based on projection with Double Nonlinear Integrals with the classical method- Naïve Bayes. The results show that our classification model is better than Naïve Bayes.