Analysing the Hierarchical Fuzzy Rule Based Classification Systems with genetic rule selection

Fernández, Alberto; Jesus, María José del; Herrera, Francisco

doi:10.1109/gefs.2010.5454155

Cited by 7 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Observing through Table II to Table VII, the classification rate and rule numbers has adequate improvements after using the algorithm proposed in this study for these three database. The result also compare to chi3, ch5, and the hierarchical fuzzy rule based classification system proposed in [2], as shown in Table VIII. From Table VIII, the simulation is better than the other works in literature.…”

Section: Resultsmentioning

confidence: 95%

“…The process to acquire adequate fuzzy rules for classification is always time consuming when the data exist in high dimension [1]. A. Fern´andez [2] proposed the idea of hierarchical fuzzy rule based classification system for classification problem. The idea to utilize hierarchical structure for the fuzzy classification model is born to assist the classification process.…”

Section: [ ]mentioning

confidence: 99%

See 1 more Smart Citation

Enhanced hierarchical fuzzy model using evolutionary GA with modified ABC algorithm for classification problem

Feng

Chiang

2015

2015 International Conference on Informative and Cybernetics for Computational Social Systems (ICCSS)

View full text Add to dashboard Cite

This paper enhances the hierarchical fuzzy model to deal with the classification problems by adopting evolutionary genetic algorithm (GA) with a modified artificial bee colony (ABC) algorithm. Traditionally, fuzzy classifier could not provide a sufficiently high classification rate in higher feature dimension with few rules. In the literature, the genetic algorithm can take advantage from the global searching; moreover, the characteristic of ABC can enhance the local searching. Therefore, the hierarchical fuzzy model integrates GA with a modified ABC algorithm is constructed in this study to recognize some classification problems. The classification simulation includes three benchmark databases such as Glass, Wine, and Iris database. The result demonstrates that using evolutionary GA and modified ABC algorithm is beneficial than that without turning. Therefore, it is clearly that our methodology considers not only the global exploration but also the local exploitation.

show abstract

Section: Resultsmentioning

confidence: 95%

Section: [ ]mentioning

confidence: 99%

Enhanced hierarchical fuzzy model using evolutionary GA with modified ABC algorithm for classification problem

Feng

Chiang

2015

2015 International Conference on Informative and Cybernetics for Computational Social Systems (ICCSS)

View full text Add to dashboard Cite

show abstract

“…In the second level, a genetic rule selection procedure is performed, which selects the best cooperating rules from the hierarchical rule base (HRB). However, the rule extraction algorithm (REA) considered in this approach results in the generation of a high number of fuzzy rules, as the results in [6] indicate. Another drawback of the approach is that it does not consider any feature reduction procedure, resulting in fuzzy rules with a complex linguistic structure when dealing with high-dimensional classification problems.…”

Section: Introductionmentioning

confidence: 95%

“…In order to increase classification accuracy, a hierarchical framework for building FRBCSs has been recently proposed [5]- [6]. This approach applies a two-level methodology.…”

Section: Introductionmentioning

confidence: 99%

A hierarchical genetic fuzzy rule-based classifier for high-dimensional classification problems

Stavrakoudis

Gitas

Theocharis

2011

2011 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE 2011)

View full text Add to dashboard Cite

This paper proposes a novel Hierarchical Genetic Fuzzy Rule-Based Classification System (HGFRBCS), targeted at effectively handling high-dimensional classification tasks. A hierarchical fuzzy rule base comprises rules with linguistic terms from a multi-granular fuzzy sets database, whereby lower levels define thicker granularities of the input space fuzzy partition. The proposed system is developed through sequential repeating steps: in each step a fuzzy rule base is created using a given granularity. Subsequently, the best performing rules are inserted in the hierarchical rule base and the process is repeated again, considering a thicker granularity. The whole process is coordinated by a boosting scheme, which localizes new rules in uncovered regions of the feature space. Comparative results for various real-world high-dimensional classification problems indicate the effectiveness of the proposed methodology.Fuzzy rule-based classification system (FRBCS); hierachical fuzzy partitions; genetic fuzzy systems

show abstract

“…Recently, a hierarchical framework for building GFRBCSs has been proposed, [19][20] which extends to classification problems the method proposed in Ref. 21 for modelling tasks.…”

Section: Introductionmentioning

confidence: 99%

Handling Highly-Dimensional Classification Tasks With Hierarchical Genetic Fuzzy Rule-Based Classifiers

Stavrakoudis

Theocharis

2012

Int. J. Unc. Fuzz. Knowl. Based Syst.

View full text Add to dashboard Cite

Many modern classification tasks are defined in highly-dimensional feature spaces. The derivation of high-performing genetic fuzzy rule-based classification systems (GFRBCSs) in such scenarios is a non-trivial task. This paper presents a framework for increasing the performance of GFRBCSs by creating a hierarchical fuzzy rule-based classifier. The proposed system is constructed through repeated invocations to a base GFRBCS procedure, considering at each step an input space fuzzy partition of a certain granularity. The best performing rules are inserted in the hierarchical rule base and the process is repeated again, considering a thicker granularity. The employed boosting scheme guides the algorithm in creating new rules to treat uncovered or misclassified patterns, thus monotonically increasing the performance of the classifier. Extensive experimental analysis in a number of real-world high-dimensional classification tasks proves the effectiveness of the proposed approach in increasing the performance of the base classifier, maintaining its interpretability to a considerable degree. are produced, even for relatively simple classification tasks, as the reported results in Ref. 20 indicate. Moreover, the lack of any feature selection procedure intensifies the problem, resulting at the same time in fuzzy rules with a complex linguistic representation when dealing with high-dimensional problems.Another hierarchical GFRBCS (HGFRBCS) has been previously proposed in Ref. 23, considering a multi-granular FSDB. In particular, an initial pool of candidate fuzzy rules is considered, defined in fuzzy partitions with different granularities. Subsequently, a genetic rule selection procedure selects the best cooperating rules, in order to form the final rule base. Although not initially proposed as a hierarchical FRBCS, the final rule base produced by this approach is a typical HRB. The only difference lies in the creation of the fuzzy rules, which are not produced iteratively, but rather selected from the initial pool of rules. This constitutes the main drawback of the methodology, as this pool is created by evaluating all possible rules for each fuzzy partition granularity, rendering the procedure practically inapplicable for high-dimensional problems.In this paper we propose the Hierarchical Rule-based Linguistic Classifier (HiRLiC), a framework for creating HGFRBCSs, aiming at producing compact yet high-performing HRBs, particularly in high-dimensional classification tasks. The basic idea is to repeatedly invoke a rule base extraction algorithm (RBEA), each time considering a fuzzy partition of thicker granularity. Fuzzy rules that increase the HRB's performance are maintained, whereas the rest are disregarded. In order for new rules to be created in uncovered regions of the feature space, we consider a boosting scheme that penalizes previously well covered patterns, so that new valuable rules will be localized in uncovered or mixed regions of the feature space. As the basic RBEA we consider a GFRBCS from a previous contri...

show abstract

Analysing the Hierarchical Fuzzy Rule Based Classification Systems with genetic rule selection

Cited by 7 publications

References 11 publications

Enhanced hierarchical fuzzy model using evolutionary GA with modified ABC algorithm for classification problem

Enhanced hierarchical fuzzy model using evolutionary GA with modified ABC algorithm for classification problem

A hierarchical genetic fuzzy rule-based classifier for high-dimensional classification problems

Handling Highly-Dimensional Classification Tasks With Hierarchical Genetic Fuzzy Rule-Based Classifiers

Contact Info

Product

Resources

About