Combining GP operators with SA search to evolve fuzzy rule based classifiers

Sánchez, Luciano; Couso, Inés; Corrales, J.A.

doi:10.1016/s0020-0255(01)00146-3

Cited by 118 publications

(73 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, in our opinion, the experimentation that support this assert was intended to solve problems based in a linear genotype, and it is not immediate to extrapolate all of their conclusions to tree-based representations. In previous works [42], we have combined a simulated annealing (SA) global search with a grammar-tree-based codification, in the context of the learning of fuzzy rules. An strategy so simple as keeping only one individual, and repeatedly mutating it, admitting or discarding the result according to a probability decreasing with time and distance, was able to improve the results of the GA. With this result in mind, in this paper we will extend our own algorithm to multiobjective problems, and propose a new population-based, multiobjective SA search (MOSA) able to elicit a set of nondominated solutions.…”

Section: Evolutionary Transparent Modeling Of Chaotic Systemsmentioning

confidence: 99%

“…The parametric crossover takes place between the parts of the individuals that derive from the production rule "Parameters", and the structural crossover involves the parts originated in the production "Structure". Leaving apart the differences in the grammar, the same operators proposed in [42] were used:…”

Section: Genetic Crossover and Mutationmentioning

confidence: 99%

“…In previous works [42], we have proposed to implement the macromutation in the SA algorithm by means of a subtree crossover with a randomly generated individual [21,37]. In our MOSA implementation we will use this technique: crossover with a random individual followed by a selection at random from the offspring.…”

Section: Genetic Crossover and Mutationmentioning

confidence: 99%

“…When vector based individuals are used, the euclidean distance can be used, but this is not longer true with tree-based representations. In previous works [42], we postulated the use of an edition distance between trees as the number of edition operations (add, remove or replace a node) needed to transform the current into the new model. Besides, in the same paper we also checked that there was possible that proximal individuals had a very different evaluation of the fitness, and the same happens here.…”

Section: The Distance Operatormentioning

confidence: 99%

See 3 more Smart Citations

Multiobjective Evolutionary Search of Difference Equations-based Models for Understanding Chaotic Systems

Sánchez

Villar

2008

Foundations of Generic Optimization

Self Cite

View full text Add to dashboard Cite

In control engineering, it is well known that many physical processes exhibit a chaotic component. In point of fact, it is also assumed that conventional modeling procedures disregard it, as stochastic noise, beside nonlinear universal approximators (like neural networks, fuzzy rule-based or genetic programming-based models,) can capture the chaotic nature of the process.In this chapter we will show that this is not always true. Despite the nonlinear capabilities of the universal approximators, these methods optimize the one step prediction of the model. This is not the most adequate objective function for a chaotic model, because there may exist many different nonchaotic processes that have near zero prediction error for such an horizon. The learning process will surely converge to one of them. Unless we include in the objective function some terms that depend on the properties on the reconstructed attractor, we may end up with a non chaotic model. Therefore, we propose to follow a multiobjective approach to model chaotic processes, and we also detail how to apply either genetic algorithms or simulated annealing to obtain a difference equations-based model.

show abstract

Section: Evolutionary Transparent Modeling Of Chaotic Systemsmentioning

confidence: 99%

Section: Genetic Crossover and Mutationmentioning

confidence: 99%

Section: Genetic Crossover and Mutationmentioning

confidence: 99%

Section: The Distance Operatormentioning

confidence: 99%

See 2 more Smart Citations

Multiobjective Evolutionary Search of Difference Equations-based Models for Understanding Chaotic Systems

Sánchez

Villar

2008

Foundations of Generic Optimization

Self Cite

View full text Add to dashboard Cite

show abstract

“…An initial paper in this topic is Fuzzy GP, developed by Geyer-Schulz [6], which combines a simple GA that operates on a contextfree language with a context-free fuzzy rule language. Sánchez et al propose an FRBCS learning process in [18] and [5] by combining GP operators with simulated annealing and GA respectively to establish the membership functions. Mendes et al develop in [14] a co-evolutionary algorithm which includes a GP based algorithm for FRBCS learning and an evolutionary algorithm to evolve membership functions.…”

Section: Introductionmentioning

confidence: 99%

A Genetic-Programming-Based Approach for the Learning of Compact Fuzzy Rule-Based Classification Systems

Berlanga

Jesús

Gacto

et al. 2006

Artificial Intelligence and Soft Computing – ICAISC 2006

View full text Add to dashboard Cite

Abstract. In the design of an interpretable fuzzy rule-based classification system (FRBCS) the precision as much as the simplicity of the extracted knowledge must be considered as objectives. In any inductive learning algorithm, when we deal with problems with a large number of features, the exponential growth of the fuzzy rule search space makes the learning process more difficult. Moreover it leads to an FRBCS with a rule base with a high cardinality. In this paper, we propose a genetic-programming-based method for the learning of an FRBCS, where disjunctive normal form (DNF) rules compete and cooperate among themselves in order to obtain an understandable and compact set of fuzzy rules, which presents a good classification performance with high dimensionality problems. This proposal uses a token competition mechanism to maintain the diversity of the population. The good results obtained with several classification problems support our proposal.

show abstract

Web usage mining for predicting final marks of students that use Moodle courses

Romero

Espejo

Zafra

et al. 2013

Comp Applic In Engineering

239

120

View full text Add to dashboard Cite

This paper shows how web usage mining can be applied in e-learning systems in order to predict the marks that university students will obtain in the final exam of a course. We have also developed a specific Moodle mining tool oriented for the use of not only experts in data mining but also of newcomers like instructors and courseware authors. The performance of different data mining techniques for classifying students are compared, starting with the student's usage data in several Cordoba University Moodle courses in engineering. Several wellknown classification methods have been used, such as statistical methods, decision trees, rule and fuzzy rule induction methods, and neural networks. We have carried out several experiments using all available and filtered data to try to obtain more accuracy. Discretization and rebalance pre-processing techniques have also been used on the original numerical data to test again if better classifier models can be obtained. Finally, we show examples of some of the models discovered and explain that a classifier model appropriate for an educational environment has to be both accurate and comprehensible in order for instructors and course administrators to be able to use it for decision making.

show abstract

Combining GP operators with SA search to evolve fuzzy rule based classifiers

Cited by 118 publications

References 11 publications

Multiobjective Evolutionary Search of Difference Equations-based Models for Understanding Chaotic Systems

Multiobjective Evolutionary Search of Difference Equations-based Models for Understanding Chaotic Systems

A Genetic-Programming-Based Approach for the Learning of Compact Fuzzy Rule-Based Classification Systems

Web usage mining for predicting final marks of students that use Moodle courses

Contact Info

Product

Resources

About