Fuzzy Rule Interpolation With $K$-Neighbors for TSK Models

When a fuzzy system is presented with an incomplete (or sparse) rule base, interpolation (FRI) offers useful mechanism to infer conclusions for unmatched observations. However, most existing FRI methodologies are established Mamdani inference models, but not Takagi–Sugeno–Kang (TSK) ones. This article presents novel approach computing interpolated outcomes TSK using only small number of neighboring rules observation. Compared methods, the new helps improve computational efficiency overall interpolative reasoning process, while minimizing adverse impact on accuracy induced by firing those low similarities For problems that involve base large size, where closest may be rather alike one another, rule-clustering-based method introduced. It derives conclusion first clustering into different groups algorithm and then, utilizing each selected from given, clusters. Systematic experimental examinations carried out verify efficacy introduced techniques, in comparison state-of-the-art over range benchmark regression problems, employing algorithms (which also shows flexibility ways implementing approach).