Selecting fuzzy if-then rules for classification problems using genetic algorithms

Ishibuchi, Hisao; Nozaki, Kengo; Yamamoto, Naohisa; Tanaka, Hideo

doi:10.1109/91.413232

Cited by 687 publications

(286 citation statements)

References 17 publications

Supporting

Mentioning

267

Contrasting

Unclassified

Order By: Relevance

“…For instance, the presence of a difficult role may alter the interpretation of a sentence and make other labeling decisions more complicated. We thus propose a fuzzy classification model with two layers (Ishibuchi et al, 1995) of SVM classifiers (Wang et al, 2016), which introduces the context of the task using fuzzy indicators to model the interplay between the two groups of features. Specifically, we train a local-layer SVM classifier L l using the sentence-level features X l (computed from sentences).…”

Section: Classification Modelmentioning

confidence: 99%

CROWD-IN-THE-LOOP: A Hybrid Approach for Annotating Semantic Roles

Wang

Akbik

Chiticariu

et al. 2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

Crowdsourcing has proven to be an effective method for generating labeled data for a range of NLP tasks. However, multiple recent attempts of using crowdsourcing to generate gold-labeled training data for semantic role labeling (SRL) reported only modest results, indicating that SRL is perhaps too difficult a task to be effectively crowdsourced. In this paper, we postulate that while producing SRL annotation does require expert involvement in general, a large subset of SRL labeling tasks is in fact appropriate for the crowd. We present a novel workflow in which we employ a classifier to identify difficult annotation tasks and route each task either to experts or crowd workers according to their difficulties. Our experimental evaluation shows that the proposed approach reduces the workload for experts by over two-thirds, and thus significantly reduces the cost of producing SRL annotation at little loss in quality.

show abstract

Section: Classification Modelmentioning

confidence: 99%

CROWD-IN-THE-LOOP: A Hybrid Approach for Annotating Semantic Roles

Wang

Akbik

Chiticariu

et al. 2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…Genetic rule selection was proposed for designing fuzzy rule-based classifiers with high accuracy and high comprehensibility in [14] where a scalar fitness function was defined as a weighted sum of two objectives: to maximize the number of correctly classified training patterns and to minimize the number of fuzzy rules. A two-objective genetic algorithm was used in [10] for finding non-dominated rule sets with respect to these two objectives.…”

Section: Heuristic Rule Extraction and Genetic Rule Selectionmentioning

confidence: 99%

Evolutionary Multiobjective Optimization for Generating an Ensemble of Fuzzy Rule-Based Classifiers

Ishibuchi

Yamamoto

2003

Genetic and Evolutionary Computation — GECCO 2003

View full text Add to dashboard Cite

Abstract. One advantage of evolutionary multiobjective optimization (EMO) algorithms over classical approaches is that many non-dominated solutions can be simultaneously obtained by their single run. In this paper, we propose an idea of using EMO algorithms for constructing an ensemble of fuzzy rule-based classifiers with high diversity. The classification of new patterns is performed based on the vote of multiple classifiers generated by a single run of EMO algorithms. Even when the classification performance of individual classifiers is not high, their ensemble often works well. The point is to generate multiple classifiers with high diversity. We demonstrate the ability of EMO algorithms to generate various non-dominated fuzzy rule-based classifiers with high diversity by their single run. Through computational experiments on some wellknown benchmark data sets, it is shown that the vote of generated fuzzy rulebased classifiers leads to high classification performance on test patterns.

show abstract

“…Several authors have proposed a genetic algorithm for fuzzy neural parameters optimization to adjust the control points of membership functions or to tune the weightings [9][10][11][12][13][14]. The pioneer was Karr [9] , who used GAs to adjust membership functions.…”

Section: Introductionmentioning

confidence: 99%

“…The pioneer was Karr [9] , who used GAs to adjust membership functions. Ishibuchi et al [10] proposed a genetic-based method for selecting a small number of significant fuzzy rules to construct a compact fuzzy classification system with high classification power. Ishibuchi and Yamamoto farther developed this idea by using mult-objective genetic local search algorithms in [13].…”

Section: Introductionmentioning

confidence: 99%