Types I & II Partition-Based Fuzzy Classifiers for Enhanced Interpretability, Smaller Output Variance and Denoising Capability

While the existing type I or II first-order Takagi-Sugeno-Kang (TSK) fuzzy classifiers generally take an additive type I or II first-order rule space followed by a non-additive defuzzification without or with type-reduction, this study enriches their family by opening a reverse thinking about a non-additive type I or II first-order rule space, followed by a simple additive defuzzification with the avoidance of explicit type-reduction calculation. Towards this end, the novel Types I & II Partition-based first-order TSK Fuzzy Classifiers (TIP-FC and TIIP-FC) and their training methods are developed for first-order linearity of consequent of each rule on all the fuzzy partitions themselves or their simple variants rather than each original feature, thereby having their distinct values in the sense of enhanced interpretability, smaller output variance, and/or denoising capability. Extensive experimental results on the adopted benchmark datasets indicate that both TIP-FC and TIIP-FC keep at least comparable to and even superior over the comparative methods in the sense of interpretability, classification performance, and/or denoising capability.