Fuzzy Information Granulation Based Decision Support Applications

Jianhong Luo,Dezhao Chen

DOI: https://doi.org/10.1109/ISIP.2008.96

2008-01-01

Abstract:Due to the learning problem on skewed distribution datasets, which tend to produce high accuracy over the majority class but poor predictive accuracy over the minority class by traditional machine learning algorithms, fuzzy information granulation based knowledge discovery and decision support model called FIG mode is proposed in this paper to improve classification performance and make effective decision support. It uses an index called ldquoSIGrdquo to select the suitable level of granularity and two membership functions to describe the features of information granules, then knowledge rules abstracted from the information granules are used to predict unknown patterns. The experimental results show that the FIG model can improve classification performance, and the performance indexes, such as G-mean, also show its better performance on skewed datasets than C4.5.