Abstract: The article presents an algorithm for automatic Mamdani-type fuzzy inference system configuration for approximation task. The constructed inference systems uses the optimal fuzzy hyperellipse-shaped clusters search and allows to clarify the inference steps. The construction is based on fuzzy Gusftaffson-Kessel clustering procedure and detects a set of clusters for all training vectors. The clustering results und later genetic algorithm processing allow to find the set of inference rules, types of fuzzy sets and all membership functions factors. The analysis of efficiency of proposed algorithm was performed with the complex surface recovery task.
Keywords: mamdani-type fuzzy inference system, fuzzy clustering, genetic algorithm
Authors and copyright holders:
—
For citation:
Silich V. A., Silich M. P., Aksyonov S. V. A Mamdani-type fuzzy system construction algorithm based on training vectors density analysis. Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki, 2013, no. 3(29), pp. 76–82.
Last issue
-2025-1.jpg)
Editorial office address
Executive Secretary of the Editor’s Office
Editor’s Office: 40 Lenina Prospect, Tomsk, 634050, Russia
Phone / Fax: + 7 (3822) 701-582
Viktor N. Maslennikov
Executive Secretary of the Editor’s Office
Editor’s Office: 40 Lenina Prospect, Tomsk, 634050, Russia
Phone / Fax: + 7 (3822) 51-21-21 / 51-43-02
Subscription for updates
