Multi-objective hierarchical genetic algorithm for interpretable fuzzy rule-based knowledge extraction

Abstract

11A new scheme based on multi-objective hierarchical genetic algorithm (MOHGA) is proposed to extract interpretable rule-based knowledge from data. The approach is derived from the use of multiple objective genetic 13 algorithm (MOGA), where the genes of the chromosome are arranged into control genes and parameter genes. These genes are in a hierarchical form so that the control genes can manipulate the parameter genes in a more 15 effective manner. The effectiveness of this chromosome formulation enables the fuzzy sets and rules to be optimally reduced. Some important concepts about the interpretability are introduced and the fitness function in the MOGA 17 will consider both the accuracy and interpretability of the fuzzy model. In order to remove the redundancy of the rule base proactively, we further apply an interpretability-driven simplification method to newborn individuals. In 19 our approach, we first apply the fuzzy clustering to generate an initial rule-based model. Then the multi-objective hierarchical genetic algorithm and the recursive least square method are used to obtain the optimized fuzzy models. 21The accuracy and the interpretability of fuzzy models derived by this approach are studied and presented in this paper. We compare our work with other methods reported in the literature on four examples: a synthetic nonlinear 23 dynamic system, a nonlinear static system, the Lorenz system and the Mackey-Glass system. Simulation results show that the proposed approach is effective and practical in knowledge extraction. 25

Improving Transparency in Approximate Fuzzy Modeling Using Multi-objective Immune-Inspired Optimisation

“…Copyright: the authors 27,28,29,30 Under this stream, a similarity measure is taken so that similar fuzzy sets can be merged. Consequently, similar rules are merged as well.…”

“…The learning procedure described in these research investigations can still be labeled as being a separate learning process so that model simplifications rely heavily on the pre-specified thresholds according to the designer's choice. Wang et al 30 proposed a hierarchical scheme to evolve both parts. However, a rule matrix was required, which rendered the scheme vulnerable to high dimensional problems due to the exponential increase in the matrix dimension.…”

Engineering Evolutionary Intelligent Systems: Methodologies, Architectures and Reviews

“…Wang et al [74] proposed a multiobjective hierarchical genetic algorithm (MOHGA) to extract interpretable rule-based knowledge from data. In order to remove the redundancy of the rule base proactively, authors applied an interpretability-driven simplification method.…”

Rule Base Reduction Using Conflicting and Reinforcement Measures

“…Several approaches for reducing fuzzy rule base have been proposed using different techniques such as interpolation methods, orthogonal transformation methods, clustering techniques [3][4][5][6][7][8]. A typical tool to perform model simplification is merging similar fuzzy sets and rules using similarity measures [9][10][11][12][13][14].…”