Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601050
Chun-Hao Chen, Ji-Syuan He, T. Hong
In this paper, we propose a two-stage multi-objective fuzzy mining algorithm for dealing with linguistic knowledge discovery. In the first stage, the multi-objective genetic algorithm is used to derive a set of non-dominated membership functions (Pareto solutions) with two objective functions. In the second stage, the clustering technique is utilized to find representative solutions from the Pareto solutions. The representative solutions could be employed to mine fuzzy association rules according to the favorites of decision makers. Experiments on a simulation dataset are made and the results show the effectiveness of the proposed algorithm.
{"title":"A two-stage multi-objective genetic-fuzzy mining algorithm","authors":"Chun-Hao Chen, Ji-Syuan He, T. Hong","doi":"10.1109/GEFS.2013.6601050","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601050","url":null,"abstract":"In this paper, we propose a two-stage multi-objective fuzzy mining algorithm for dealing with linguistic knowledge discovery. In the first stage, the multi-objective genetic algorithm is used to derive a set of non-dominated membership functions (Pareto solutions) with two objective functions. In the second stage, the clustering technique is utilized to find representative solutions from the Pareto solutions. The representative solutions could be employed to mine fuzzy association rules according to the favorites of decision makers. Experiments on a simulation dataset are made and the results show the effectiveness of the proposed algorithm.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127465239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601053
Dalai Tang, János Botzheim, N. Kubota, Toru Yamaguchi
This paper analyzes the performance of human transport mode estimation by fuzzy spiking neural network in informationally structured space based on smart phone sensor. The importance of information structuralization is considered. In our previous work we applied spiking neural network to extract the human position in a room equipped with sensor network devices. In this paper fuzzy spiking neural network is applied to extract the human activity outdoors when equipped with smart phone sensor. We discuss how to update the base value by preprocessing for generating the input values to the spiking neurons. The learning method of the spiking neural network based on the time series of the measured data is explained as well. Evolution strategy is used for optimizing the parameters of the fuzzy spiking neural network. Several experimental results are presented for confirming the effectiveness of the proposed method.
{"title":"Estimation of human transport modes by fuzzy spiking neural network and evolution strategy in informationally structured space","authors":"Dalai Tang, János Botzheim, N. Kubota, Toru Yamaguchi","doi":"10.1109/GEFS.2013.6601053","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601053","url":null,"abstract":"This paper analyzes the performance of human transport mode estimation by fuzzy spiking neural network in informationally structured space based on smart phone sensor. The importance of information structuralization is considered. In our previous work we applied spiking neural network to extract the human position in a room equipped with sensor network devices. In this paper fuzzy spiking neural network is applied to extract the human activity outdoors when equipped with smart phone sensor. We discuss how to update the base value by preprocessing for generating the input values to the spiking neurons. The learning method of the spiking neural network based on the time series of the measured data is explained as well. Evolution strategy is used for optimizing the parameters of the fuzzy spiking neural network. Several experimental results are presented for confirming the effectiveness of the proposed method.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130537652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601048
Yi-Ling Liu, F. Gomide
Genetic Fuzzy Systems have been successfully used as a modeling approach for numerous applications. There is an increasing interest on how to construct fuzzy models for different types of complex systems such as highly nonlinear, large-scale, multiobjective, and high-dimensional systems. Current state of the art indicates the use of fast and scalable evolutionary algorithms in complex fuzzy modeling tasks. Genetic fuzzy systems offer an effective approach to embed genetic database learning and fast learning of parsimonious and accurate models. This paper suggests a participatory genetic learning approach as a tool for genetic fuzzy system modeling. Participatory genetic learning is an evolutionary computation paradigm in which the population itself plays an important role to assign fitness values to individuals. The approach uses compatibility between two randomly chosen individuals and the fittest to select the mates, and selective transfer recombination mechanism to exchange information between mates. Mutation is done similarly as in the canonical genetic algorithm. The usage of participatory learning, selective transfer, and mutation translates into a new type of genetic algorithm for genetic fuzzy system modeling. This paper focuses on the application of participatory genetic learning for rule-based fuzzy modeling of regression problems. Actual data concerning an electric system maintenance problem and results reported in the literature are employed to evaluate the performance of participatory genetic learning. The mean squared error and number of rules measure modeling accuracy and complexity, respectively. The result shows that participatory genetic learning produces accurate, parsimonious models, and is fast when compared with current state of the art approaches.
{"title":"Participatory genetic learning in fuzzy system modeling","authors":"Yi-Ling Liu, F. Gomide","doi":"10.1109/GEFS.2013.6601048","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601048","url":null,"abstract":"Genetic Fuzzy Systems have been successfully used as a modeling approach for numerous applications. There is an increasing interest on how to construct fuzzy models for different types of complex systems such as highly nonlinear, large-scale, multiobjective, and high-dimensional systems. Current state of the art indicates the use of fast and scalable evolutionary algorithms in complex fuzzy modeling tasks. Genetic fuzzy systems offer an effective approach to embed genetic database learning and fast learning of parsimonious and accurate models. This paper suggests a participatory genetic learning approach as a tool for genetic fuzzy system modeling. Participatory genetic learning is an evolutionary computation paradigm in which the population itself plays an important role to assign fitness values to individuals. The approach uses compatibility between two randomly chosen individuals and the fittest to select the mates, and selective transfer recombination mechanism to exchange information between mates. Mutation is done similarly as in the canonical genetic algorithm. The usage of participatory learning, selective transfer, and mutation translates into a new type of genetic algorithm for genetic fuzzy system modeling. This paper focuses on the application of participatory genetic learning for rule-based fuzzy modeling of regression problems. Actual data concerning an electric system maintenance problem and results reported in the literature are employed to evaluate the performance of participatory genetic learning. The mean squared error and number of rules measure modeling accuracy and complexity, respectively. The result shows that participatory genetic learning produces accurate, parsimonious models, and is fast when compared with current state of the art approaches.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132685938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601052
Ana M. Palacios, L. Sánchez, Inés Couso
Existing extensions of AdaBoost-based fuzzy rule learning to low quality databases yield suboptimal results in multi-class problems. A new procedure is proposed where the original multi-class database is transformed into several multi-label problems that can be tackled with binary AdaBoost. The performance of this proposal is assessed in comparison with other classification schemes for imprecise data. A novel experimental design for imprecise databases is introduced for this last purpose. The new algorithm is applied to a set of real-world and synthetic low quality datasets.
{"title":"Boosting fuzzy rules with low quality data in multi-class problems: Open problems and challenges","authors":"Ana M. Palacios, L. Sánchez, Inés Couso","doi":"10.1109/GEFS.2013.6601052","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601052","url":null,"abstract":"Existing extensions of AdaBoost-based fuzzy rule learning to low quality databases yield suboptimal results in multi-class problems. A new procedure is proposed where the original multi-class database is transformed into several multi-label problems that can be tackled with binary AdaBoost. The performance of this proposal is assessed in comparison with other classification schemes for imprecise data. A novel experimental design for imprecise databases is introduced for this last purpose. The new algorithm is applied to a set of real-world and synthetic low quality datasets.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128818534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601051
S. Fukuda
Uncertainties originating from observation data and modelling approaches can affect model accuracy and thus impact on the applicability and reliability of a model. This paper aims to assess the effects of data prevalence (i.e., proportion of presence in the entire data set) on species distribution modelling and habitat preference evaluation using a 0-order genetic Takagi-Sugeno fuzzy model. The effects were evaluated based on the model accuracy and habitat preference curves (HPCs). In order to avoid the data uncertainty, virtual species data were generated using hypothetical HPCs under different assumptions on the interaction between habitat variables and habitat preference of a virtual fish. In total, thirteen data sets under three different interaction scenarios were generated. The model accuracy of resulting models was different according to the data prevalence, whereas different trends between data sets under different interaction scenarios were observed. Although the HPC shapes were similar across data sets, the HPCs were different according to the data prevalence, of which a higher prevalence can result in a uniform HPC. This study demonstrates possible influences of data prevalence on the species distribution modelling. Further study is needed for a better solution to cope with the prevalence-related problems in ecological modelling.
{"title":"Effects of data prevalence on species distribution modelling using a genetic takagi-sugeno fuzzy system","authors":"S. Fukuda","doi":"10.1109/GEFS.2013.6601051","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601051","url":null,"abstract":"Uncertainties originating from observation data and modelling approaches can affect model accuracy and thus impact on the applicability and reliability of a model. This paper aims to assess the effects of data prevalence (i.e., proportion of presence in the entire data set) on species distribution modelling and habitat preference evaluation using a 0-order genetic Takagi-Sugeno fuzzy model. The effects were evaluated based on the model accuracy and habitat preference curves (HPCs). In order to avoid the data uncertainty, virtual species data were generated using hypothetical HPCs under different assumptions on the interaction between habitat variables and habitat preference of a virtual fish. In total, thirteen data sets under three different interaction scenarios were generated. The model accuracy of resulting models was different according to the data prevalence, whereas different trends between data sets under different interaction scenarios were observed. Although the HPC shapes were similar across data sets, the HPCs were different according to the data prevalence, of which a higher prevalence can result in a uniform HPC. This study demonstrates possible influences of data prevalence on the species distribution modelling. Further study is needed for a better solution to cope with the prevalence-related problems in ecological modelling.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127700071","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601054
Michela Fazzolari, R. Alcalá, Y. Nojima, H. Ishibuchi, F. Herrera
A multi-objective evolutionary fuzzy rule selection process extracts a subset of fuzzy rules from an initial set, by applying a multi-objective evolutionary algorithm. Two approaches can be used to determine the number of terms (i.e. the granularity) associated with the linguistic variables that appear in the rules: a pre-established single granularity can be chosen, or a multiple granularities approach can be preferred. The latter favors a reduction in the number of extracted rules, but it also brings to a possible loss of interpretability. To prevent this problem, suitable granularities can be determined by applying automatic techniques before the initial rule generation process. In this contribution, we investigate how the application of a single granularity learning approach influences the performance of fuzzy associative rule-based classifiers. The aim is to reduce the complexity of the obtained models, trying to maintain a good classification ability.
{"title":"Improving a fuzzy association rule-based classification model by granularity learning based on heuristic measures over multiple granularities","authors":"Michela Fazzolari, R. Alcalá, Y. Nojima, H. Ishibuchi, F. Herrera","doi":"10.1109/GEFS.2013.6601054","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601054","url":null,"abstract":"A multi-objective evolutionary fuzzy rule selection process extracts a subset of fuzzy rules from an initial set, by applying a multi-objective evolutionary algorithm. Two approaches can be used to determine the number of terms (i.e. the granularity) associated with the linguistic variables that appear in the rules: a pre-established single granularity can be chosen, or a multiple granularities approach can be preferred. The latter favors a reduction in the number of extracted rules, but it also brings to a possible loss of interpretability. To prevent this problem, suitable granularities can be determined by applying automatic techniques before the initial rule generation process. In this contribution, we investigate how the application of a single granularity learning approach influences the performance of fuzzy associative rule-based classifiers. The aim is to reduce the complexity of the obtained models, trying to maintain a good classification ability.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133781080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601056
Y. Nojima, H. Ishibuchi
Genetic fuzzy rule selection has been frequently used for fuzzy rule-based classifier design. A number of its variants have also been proposed in the literature. In many studies on genetic fuzzy rule selection, each antecedent condition in fuzzy rules is given for a single input variable such as “x1 is small” and “x2 is large”. As a result, each antecedent fuzzy set is defined on a single input variable. In this paper, we examine the use of fuzzy relational conditions with respect to the relation between two input variables such as “x1 is approximately equal to x2” and “x3 is approximately larger than x4”. Such a fuzzy relational condition is defined by a fuzzy set on a pair of input variables. We examine the effect of using fuzzy rules with fuzzy relational conditions on the performance of fuzzy rule-based classifiers designed by multiobjective genetic fuzzy rule selection.
{"title":"Multiobjective genetic fuzzy rule selection with fuzzy relational rules","authors":"Y. Nojima, H. Ishibuchi","doi":"10.1109/GEFS.2013.6601056","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601056","url":null,"abstract":"Genetic fuzzy rule selection has been frequently used for fuzzy rule-based classifier design. A number of its variants have also been proposed in the literature. In many studies on genetic fuzzy rule selection, each antecedent condition in fuzzy rules is given for a single input variable such as “x<sub>1</sub> is small” and “x<sub>2</sub> is large”. As a result, each antecedent fuzzy set is defined on a single input variable. In this paper, we examine the use of fuzzy relational conditions with respect to the relation between two input variables such as “x<sub>1</sub> is approximately equal to x<sub>2</sub>” and “x<sub>3</sub> is approximately larger than x<sub>4</sub>”. Such a fuzzy relational condition is defined by a fuzzy set on a pair of input variables. We examine the effect of using fuzzy rules with fuzzy relational conditions on the performance of fuzzy rule-based classifiers designed by multiobjective genetic fuzzy rule selection.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131649666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601049
David García, A. G. Muñoz, Raúl Pérez
The main difficulty faced by a learning algorithm is to find the appropriate knowledge inside of the huge search space of possible solutions. Typically, the researchers try to solve this problem developing more efficient search algorithms, defining “ad-hoc” heuristic for the specific problem or reducing the expressiveness of the knowledge representation. This work explores an alternative way that consists of reducing the search space using a completeness condition. The proposed model is implemented on NSLV, a fuzzy rule learning algorithm based on genetic algorithms. We present an experimental study of the behavior of NSLV on pruned search spaces. The experimental results show that when we work with these spaces it is possible to find a good trace-off among prediction capacity, complexity of the knowledge obtained and learning time.
{"title":"An empirical study about the behavior of a genetic learning algorithm on searching spaces pruned by a completeness condition","authors":"David García, A. G. Muñoz, Raúl Pérez","doi":"10.1109/GEFS.2013.6601049","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601049","url":null,"abstract":"The main difficulty faced by a learning algorithm is to find the appropriate knowledge inside of the huge search space of possible solutions. Typically, the researchers try to solve this problem developing more efficient search algorithms, defining “ad-hoc” heuristic for the specific problem or reducing the expressiveness of the knowledge representation. This work explores an alternative way that consists of reducing the search space using a completeness condition. The proposed model is implemented on NSLV, a fuzzy rule learning algorithm based on genetic algorithms. We present an experimental study of the behavior of NSLV on pruned search spaces. The experimental results show that when we work with these spaces it is possible to find a good trace-off among prediction capacity, complexity of the knowledge obtained and learning time.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127772122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-04-16DOI: 10.1109/GEFS.2013.6601055
Tim Brys, Mădălina M. Drugan, P. Bosman, M. D. Cock, A. Nowé
Satisfiability solving in fuzzy logics is a subject that has not been researched much, certainly compared to satisfiability in propositional logics. Yet, fuzzy logics are a powerful tool for modelling complex problems. Recently, we proposed an optimization approach to solving satisfiability in fuzzy logics and compared the standard Covariance Matrix Adaptation Evolution Strategy algorithm (CMA-ES) with an analytical solver on a set of benchmark problems. Especially on more finegrained problems did CMA-ES compare favourably to the analytical approach. In this paper, we evaluate two types of hillclimber in addition to CMA-ES, as well as restart strategies for these algorithms. Our results show that a population-based hillclimber outperforms CMA-ES on the harder problem class.
{"title":"Local search and restart strategies for satisfiability solving in fuzzy logics","authors":"Tim Brys, Mădălina M. Drugan, P. Bosman, M. D. Cock, A. Nowé","doi":"10.1109/GEFS.2013.6601055","DOIUrl":"https://doi.org/10.1109/GEFS.2013.6601055","url":null,"abstract":"Satisfiability solving in fuzzy logics is a subject that has not been researched much, certainly compared to satisfiability in propositional logics. Yet, fuzzy logics are a powerful tool for modelling complex problems. Recently, we proposed an optimization approach to solving satisfiability in fuzzy logics and compared the standard Covariance Matrix Adaptation Evolution Strategy algorithm (CMA-ES) with an analytical solver on a set of benchmark problems. Especially on more finegrained problems did CMA-ES compare favourably to the analytical approach. In this paper, we evaluate two types of hillclimber in addition to CMA-ES, as well as restart strategies for these algorithms. Our results show that a population-based hillclimber outperforms CMA-ES on the harder problem class.","PeriodicalId":362308,"journal":{"name":"2013 IEEE International Workshop on Genetic and Evolutionary Fuzzy Systems (GEFS)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131515696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：