Fuzzy rule extraction by bacterial memetic algorithms

2010 11th International Symposium on Computational Intelligence and Informatics (CINTI), 2010

Evolutionary algorithms are an important branch of soft computing, being able to provide approximate solutions to problems in a reasonable amount of time. The underlying principle can be realized in an almost unlimited number of ways. This paper presents four main variants of evolutionary algorithms, and a method of running them in a topology consisting of multiple populations. The resources given to each population and migration are altered dynamically throughout the test, based on the effectiveness they show. Along with evolutionary methods, the solutions are also adjusted by gradient-based numerical optimization, in our case the Levenberg-Marquardt algorithm. These steps are added to the evolutionary processes as an extension, resulting in what are called memetic algorithms. The specific application for these methods here is optimizing fuzzy rule-bases, thereby making inference systems better at emulating a desired behavior, such as modeling a certain objective function. I.

2016

— The process of fuzzy modeling or fuzzy model identification is an arduous task. This paper presents the application of Memetic algorithms (MAs) for the identification of complete fuzzy model that includes membership function design for input and output variables and rulebase generation from the numerical data set. We have applied the algorithms on four bench mark data: A rapid Ni-Cd battery charger, the Box & Jenkins‟s gas-furnace data, the Iris data classification problem and the wine data classification problem. The comparison of obtained results from MAs with Genetic algorithms (GAs) brings out the remarkable efficiency of MAs. The result suggests that for these problems the proposed approach is better than those suggested in the literature.

This paper presents our approach to the rule extraction problem from trained neural network. A method called REX is briefly described. REX acquires a set of fuzzy rules using an evolutionary algorithm. Evolutionary algorithm searches not only fuzzy rules, but also a description of fuzzy sets. The way of coding and evaluation process of an individual is presented. The method was tested using the following benchmark data sets: IRIS, WINE and Wisconsin Breast Cancer Diagnosis. On the basis of the experimental studies shown in this paper, we can conclude that rules obtained by REX can be easily understood by human - they include small number of premises, and their fidelity is very high. Obtained results are compared to other rule extraction methods.

for BNN design and the other is the Bacterial Evolutionary Algorithm (BEA) applied for fuzzy rule extraction. Also, the facility to incorporate a multi-objective approach to the GP algorithm is outlined, enabling the designer to obtain models more adequate for their intended use.

An algorithm is presented that uses evolutionary programming to construct fuzzy membership functions that are used to extract Zadeh-Mamdani fuzzy rules from a constructive neural network. The algorithm has potential applications in fields such as data mining and knowledge-based decision support systems. Evaluation of the algorithm over two well known benchmark data sets shows that while the results are promising, some problems are apparent. These problems provide avenues for further research.

International Conference on Fuzzy Systems, 2010

In this paper a method is proposed for constructing hierarchical fuzzy rule bases in order to model black box systems defined by input-output pairs, i.e. to solve supervised machine learning problems. The resultant hierarchical rule base is the knowledge base, which is constructed by using structure constructing evolutionary techniques, namely, Genetic and Bacterial Programming Algorithms. Applying hierarchical fuzzy rule bases is a way of reducing the complexity of the knowledge base, whereas evolutionary methods ensure a relatively efficient learning process. This is the reason of the investigation of this combination.

This paper presents the analysis of different population based algorithms for the rulebase generation from numerical data sets. As fuzzy rulebase generation is one of the key issues in fuzzy modeling. The algorithms are applied on a rapid Ni-Cd battery charger data set. In this paper, we compare the efficiency of different algorithms and conclude that SCA algorithms with local search give remarkable efficiency as compared to SCA algorithms alone. Also found that the efficiency of SCA with local search is comparable to memetic algorithms.

This paper presents the analysis of different population based algorithms for the rulebase generation from numerical data sets. As fuzzy rulebase generation is one of the key issues in fuzzy modeling. The algorithms are applied on a rapid Ni-Cd battery charger data set. In this paper, we compare the efficiency of different algorithms and conclude that SCA algorithms with local search give remarkable efficiency as compared to SCA algorithms alone. Also found that the efficiency of SCA with local search is comparable to memetic algorithms.

Proceedings of the 15th IFAC World Congress, 2002, 2002

Genetic Algorithms (GA's) and others evolutionary optimization methods to design fuzzy models from data have received much attention in recent literature. This paper presents a methodology to simultaneous design of membership functions and rules sets for fuzzy logic models. The validity of the proposed identification method has been demonstrated by simulation example of inverted pendulum.

International journal of science and research , 2025

Modeling real-world situations has made use of systems for fuzzy inference that incorporate fuzzy rule bases (FRB). One of the limitations of these traditional fuzzy inference systems is the sheer amount of fuzzy procedures and operators that an expert needs to understand. In this work, we suggest an alternative schema for learning and reasoning that is based on fuzzy functions rather than if-then rule foundation structures. A novel fuzzy functions method optimized with biological algorithms is proposed to replace the fuzzy regulators and processes used by FRBs and increase the correctness of fuzzy models. The Improved Fuzzy Clustering (IFC) approach, a guided hybrid fuzzy clustering method that yields higher membership values, serves as the foundation for the new method's architecture identification. When creating fuzzy functions and improving them through evolutionary techniques, the proposed fuzzy functions methodology has the benefit of employing values for membership and other ambiguous details regarding the natural group of data samples as additional predictors. Comparison experiments utilizing real industrial and financial data demonstrate that the proposed approach is comparable or better in the modeling of regression issue domains.