Performance Evaluation of SIRMs Models for Classification Problems
The performance of Single-Input Rule-Modules (SIRMs) models are studied for classification proglems. In the original version of SIRMs models, each fuzzy if-then rule has a single real-valued output. The final output from an SIRMs model is discretized in application to classification problems. This paper proposes an extention to the SIRMs models where each fuzzy if-then rule has multiple realvalued outputs that represent the activation level of the corresponding classes. The classification performance of both the original and the extended models are evaluated through a series of computational experiments using two-dimensional pattern classification problems.
Unable to display preview. Download preview PDF.
- 1.Yubazaki, N.: Sirms dynamically connected fuzzy inference model and its applications. In: Proc. IFSA 1997, Prague, Czech, vol. 3, pp. 410–415 (1997), http://ci.nii.ac.jp/naid/10018385046/en/
- 5.Yi, J., Yubazaki, N., Hirota, K.: A proposal of sirms dynamically connected fuzzy inference model for plural input fuzzy control. Fuzzy Sets and Systems 125(1), 79–92 (2002), http://www.sciencedirect.com/science/article/pii/S0165011400001354, doi:10.1016/S0165-0114(00)00135-4MathSciNetMATHCrossRefGoogle Scholar