Pub Date : 2006-11-30DOI: 10.1109/ISEFS.2006.251150
J. Kacprzyk, A. Wilbik, S. Zadrozny
The purpose of this paper is to propose a new easily implementable approach to a linguistic summarization of trends that may occur in temporal data, to be more specific - time series. To characterize the trends in time series, we use three parameters: dynamics of change, duration and variability, and apply to them the fuzzy linguistic summaries of data (databases) in the sense of Yager (cf. Yager (1982), Kacprzyk and Yager (2001) and Kacprzyk et al. (2000)) which in the form of natural language-like sentences subsume the very essence of a set of data. A genetic algorithm is used to generate the linguistic summaries sought
{"title":"Using a Genetic Algorithm to Derive a Linguistic Summary of Trends in Numerical Time Series","authors":"J. Kacprzyk, A. Wilbik, S. Zadrozny","doi":"10.1109/ISEFS.2006.251150","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251150","url":null,"abstract":"The purpose of this paper is to propose a new easily implementable approach to a linguistic summarization of trends that may occur in temporal data, to be more specific - time series. To characterize the trends in time series, we use three parameters: dynamics of change, duration and variability, and apply to them the fuzzy linguistic summaries of data (databases) in the sense of Yager (cf. Yager (1982), Kacprzyk and Yager (2001) and Kacprzyk et al. (2000)) which in the form of natural language-like sentences subsume the very essence of a set of data. A genetic algorithm is used to generate the linguistic summaries sought","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133951508","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251142
A. Tsakonas, N. Ampazis, G. Dounias
The paper presents methods for classification of applicants into different categories of credit risk using four different computational intelligence techniques. The selected methodologies involved in the rule-based categorization task are (1) feedforward neural networks trained with second order methods (2) inductive machine learning, (3) hierarchical decision trees produced by grammar-guided genetic programming and (4) fuzzy rule based systems produced by grammar-guided genetic programming. The data used are both numerical and linguistic in nature and they represent a real-world problem, that of deciding whether a loan should be granted or not, in respect to financial details of customers applying for that loan, to a specific private EU bank. We examine the proposed classification models with a sample of enterprises that applied for a loan, each of which is described by financial decision variables (ratios), and classified to one of the four predetermined classes. Attention is given to the comprehensibility and the ease of use for the acquired decision models. Results show that the application of the proposed methods can make the classification task easier and - in some cases - may minimize significantly the amount of required credit data. We consider that these methodologies may also give the chance for the extraction of a comprehensible credit management model or even the incorporation of a related decision support system in banking
{"title":"Towards a Comprehensible and Accurate Credit Management Model: Application of Four Computational Intelligence Methodologies","authors":"A. Tsakonas, N. Ampazis, G. Dounias","doi":"10.1109/ISEFS.2006.251142","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251142","url":null,"abstract":"The paper presents methods for classification of applicants into different categories of credit risk using four different computational intelligence techniques. The selected methodologies involved in the rule-based categorization task are (1) feedforward neural networks trained with second order methods (2) inductive machine learning, (3) hierarchical decision trees produced by grammar-guided genetic programming and (4) fuzzy rule based systems produced by grammar-guided genetic programming. The data used are both numerical and linguistic in nature and they represent a real-world problem, that of deciding whether a loan should be granted or not, in respect to financial details of customers applying for that loan, to a specific private EU bank. We examine the proposed classification models with a sample of enterprises that applied for a loan, each of which is described by financial decision variables (ratios), and classified to one of the four predetermined classes. Attention is given to the comprehensibility and the ease of use for the acquired decision models. Results show that the application of the proposed methods can make the classification task easier and - in some cases - may minimize significantly the amount of required credit data. We consider that these methodologies may also give the chance for the extraction of a comprehensible credit management model or even the incorporation of a related decision support system in banking","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126933785","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251148
Y. Nojima, H. Ishibuchi, I. Kuwajima
We developed two GA-based schemes for the design of fuzzy rule-based classification systems. One is genetic rule selection and the other is genetics-based machine learning (GBML). In our genetic rule selection scheme, first a large number of promising fuzzy rules are extracted from numerical data in a heuristic manner as candidate rules. Then a genetic algorithm is used to select a small number of fuzzy rules. A rule set is represented by a binary string whose length is equal to the number of candidate rules. On the other hand, a fuzzy rule is denoted by its antecedent fuzzy sets as an integer substring in our GBML scheme. A rule set is represented by a concatenated integer string. In this paper, we compare these two schemes in terms of their search ability to efficiently find compact fuzzy rule-based classification systems with high accuracy. The main difference between these two schemes is that GBML has a huge search space consisting of all combinations of possible fuzzy rules while genetic rule selection has a much smaller search space with only candidate rules
{"title":"Comparison of Search Ability between Genetic Fuzzy Rule Selection and Fuzzy Genetics-Based Machine Learning","authors":"Y. Nojima, H. Ishibuchi, I. Kuwajima","doi":"10.1109/ISEFS.2006.251148","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251148","url":null,"abstract":"We developed two GA-based schemes for the design of fuzzy rule-based classification systems. One is genetic rule selection and the other is genetics-based machine learning (GBML). In our genetic rule selection scheme, first a large number of promising fuzzy rules are extracted from numerical data in a heuristic manner as candidate rules. Then a genetic algorithm is used to select a small number of fuzzy rules. A rule set is represented by a binary string whose length is equal to the number of candidate rules. On the other hand, a fuzzy rule is denoted by its antecedent fuzzy sets as an integer substring in our GBML scheme. A rule set is represented by a concatenated integer string. In this paper, we compare these two schemes in terms of their search ability to efficiently find compact fuzzy rule-based classification systems with high accuracy. The main difference between these two schemes is that GBML has a huge search space consisting of all combinations of possible fuzzy rules while genetic rule selection has a much smaller search space with only candidate rules","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113983790","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251184
R. Yager
We discuss two technologies that allow the construction of intelligent systems that can evolve and learn. The first is the Hierarchical Prioritized Structure and the second the participatory learning paradigm.
{"title":"Learning Methods for Intelligent Evolving Systems","authors":"R. Yager","doi":"10.1109/ISEFS.2006.251184","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251184","url":null,"abstract":"We discuss two technologies that allow the construction of intelligent systems that can evolve and learn. The first is the Hierarchical Prioritized Structure and the second the participatory learning paradigm.","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114863064","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251179
J. C. Vidal, M. Mucientes, Alberto Bugarín-Diz, M. Lama
This paper describes an adaptive evolutionary approach to the problem of the production planning task in the wood furniture industry. The objective is to schedule new incoming orders and to regenerate the scheduling for already existing orders when necessary. Complexity and uncertainty of this task promotes the use of an hybrid solution that combines evolutionary algorithms (EAs) and fuzzy sets. On one hand, EAs allow an efficient and flexible use of large number of parameters involved in the scheduling task and to reduce its computation time. On the other hand, fuzzy sets improve the confidence in the evaluation of the solutions when uncertain knowledge is used. This evolutionary approach to the production planning task is a part of a knowledge-based system that manages the product design life cycle of wood-based furniture and is being currently implemented on a wood furniture industry
{"title":"An Adaptive Evolutionary Algorithm for Production Planning in Wood Furniture Industry","authors":"J. C. Vidal, M. Mucientes, Alberto Bugarín-Diz, M. Lama","doi":"10.1109/ISEFS.2006.251179","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251179","url":null,"abstract":"This paper describes an adaptive evolutionary approach to the problem of the production planning task in the wood furniture industry. The objective is to schedule new incoming orders and to regenerate the scheduling for already existing orders when necessary. Complexity and uncertainty of this task promotes the use of an hybrid solution that combines evolutionary algorithms (EAs) and fuzzy sets. On one hand, EAs allow an efficient and flexible use of large number of parameters involved in the scheduling task and to reduce its computation time. On the other hand, fuzzy sets improve the confidence in the evaluation of the solutions when uncertain knowledge is used. This evolutionary approach to the production planning task is a part of a knowledge-based system that manages the product design life cycle of wood-based furniture and is being currently implemented on a wood furniture industry","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124083013","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251144
Zhenyu Wang, Vasile Palade, Yong Xu
A neuro-fuzzy ensemble model (NFE) is proposed in this paper for analysing the gene expression data from microarray experiments. The proposed approach was tested on three benchmark cancer gene expression data sets. Experimental results show that our NFE model can be used as an efficient computational tool for microarray data analysis. In addition, compared to some most widely used approaches, neuro-fuzzy (NF)-based models not only supply good classification results, but their behavior can also be explained and interpreted in human understandable terms, which provides the researchers with a better understanding of the data
{"title":"Neuro-Fuzzy Ensemble Approach for Microarray Cancer Gene Expression Data Analysis","authors":"Zhenyu Wang, Vasile Palade, Yong Xu","doi":"10.1109/ISEFS.2006.251144","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251144","url":null,"abstract":"A neuro-fuzzy ensemble model (NFE) is proposed in this paper for analysing the gene expression data from microarray experiments. The proposed approach was tested on three benchmark cancer gene expression data sets. Experimental results show that our NFE model can be used as an efficient computational tool for microarray data analysis. In addition, compared to some most widely used approaches, neuro-fuzzy (NF)-based models not only supply good classification results, but their behavior can also be explained and interpreted in human understandable terms, which provides the researchers with a better understanding of the data","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121316838","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251165
Nikola Kasabov
This paper discusses opportunities and challenges for the creation of evolving artificial neural network (ANN) and more general computational intelligence (CI) models inspired by principles at different levels of information processing in the brain - neuronal-, genetic-, and quantum - and mainly the issues related to the integration of these principles into more powerful and accurate ANN models. A particular type of ANN, evolving connectionist systems (ECOS), is used to illustrate this approach. ECOS evolve their structure and functionality through continuous learning from data and facilitate data and knowledge integration and knowledge elucidation. ECOS gain inspiration from the evolving processes in the brain. Evolving fuzzy neural networks and evolving spiking neural networks are presented as examples. With more genetic information available now, it becomes possible to integrate the gene and the neuronal information into neuro-genetic models and to use them for a better understanding of complex brain processes. Further down in the information processing hierarchy are the quantum processes. Quantum inspired ANN may help solve efficiently the hardest computational problems. It may be possible to integrate quantum principles into brain-gene inspired ANN models for a faster and more accurate modeling. All the topics above are illustrated with some contemporary solutions, but many more open questions and challenges are raised and directions for further research outlined
{"title":"Neuro-, Genetic-, and Quantum Inspired Evolving Intelligent Systems","authors":"Nikola Kasabov","doi":"10.1109/ISEFS.2006.251165","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251165","url":null,"abstract":"This paper discusses opportunities and challenges for the creation of evolving artificial neural network (ANN) and more general computational intelligence (CI) models inspired by principles at different levels of information processing in the brain - neuronal-, genetic-, and quantum - and mainly the issues related to the integration of these principles into more powerful and accurate ANN models. A particular type of ANN, evolving connectionist systems (ECOS), is used to illustrate this approach. ECOS evolve their structure and functionality through continuous learning from data and facilitate data and knowledge integration and knowledge elucidation. ECOS gain inspiration from the evolving processes in the brain. Evolving fuzzy neural networks and evolving spiking neural networks are presented as examples. With more genetic information available now, it becomes possible to integrate the gene and the neuronal information into neuro-genetic models and to use them for a better understanding of complex brain processes. Further down in the information processing hierarchy are the quantum processes. Quantum inspired ANN may help solve efficiently the hardest computational problems. It may be possible to integrate quantum principles into brain-gene inspired ANN models for a faster and more accurate modeling. All the topics above are illustrated with some contemporary solutions, but many more open questions and challenges are raised and directions for further research outlined","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128146404","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251178
O. Georgieva, F. Klawonn
Following the idea to search for just one cluster at a time a prototype-based clustering algorithm named dynamic data assigning assessment (DDAA) was recently proposed. It is based on the noise clustering technique and finds single good clusters one by one and at the same time it separates the noise data. In this paper we present the basic idea and executive procedures of evolving variant of DDAA algorithm that are capable to deal with the currently entered system information. The evolving DDAA algorithm assigns every new data point to an already determined good cluster or, alternatively, to the noise cluster. It checks whether the new data collection provides a new good cluster(s) and thus, changes the data structure. The assignment could be done in hard or fuzzy sense
{"title":"Evolving Clustering via the Dynamic Data Assigning Assessment Algorithm","authors":"O. Georgieva, F. Klawonn","doi":"10.1109/ISEFS.2006.251178","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251178","url":null,"abstract":"Following the idea to search for just one cluster at a time a prototype-based clustering algorithm named dynamic data assigning assessment (DDAA) was recently proposed. It is based on the noise clustering technique and finds single good clusters one by one and at the same time it separates the noise data. In this paper we present the basic idea and executive procedures of evolving variant of DDAA algorithm that are capable to deal with the currently entered system information. The evolving DDAA algorithm assigns every new data point to an already determined good cluster or, alternatively, to the noise cluster. It checks whether the new data collection provides a new good cluster(s) and thus, changes the data structure. The assignment could be done in hard or fuzzy sense","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127262681","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251147
G. Serra, C. Bottura
This paper presents an intelligent gain scheduling adaptive control approach for nonlinear plants. A fuzzy PI discrete controller is optimally designed by using a multiobjective genetic algorithm for simultaneously satisfying the following specifications: overshoot and settling time minimizations and output response smoothing. A neural gain scheduler is designed, by the backpropagation algorithm, to tune the optimal parameters of the fuzzy PI controller at some operating points. Simulation results are shown for adaptive speed control of a DC servomotor used as actuator of robotic manipulators
{"title":"Genetic Approach for Neural Scheduling of Multiobjective Fuzzy PI Controllers","authors":"G. Serra, C. Bottura","doi":"10.1109/ISEFS.2006.251147","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251147","url":null,"abstract":"This paper presents an intelligent gain scheduling adaptive control approach for nonlinear plants. A fuzzy PI discrete controller is optimally designed by using a multiobjective genetic algorithm for simultaneously satisfying the following specifications: overshoot and settling time minimizations and output response smoothing. A neural gain scheduler is designed, by the backpropagation algorithm, to tune the optimal parameters of the fuzzy PI controller at some operating points. Simulation results are shown for adaptive speed control of a DC servomotor used as actuator of robotic manipulators","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130884457","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 : 2006-11-30DOI: 10.1109/ISEFS.2006.251133
R. Precup, S. Preitl
This paper proposes an original iterative feedback tuning (IFT) method employing genetic algorithms to develop a class of fuzzy control systems. The approach is based on using the linear case results from the original IFT method and on replacing the parameter update law by genetic algorithms. Then, these results are transferred to the fuzzy case in terms of the modal equivalence principle resulting in an attractive development method referred to as genetic iterative feedback tuning (GIFT). The GIFT method is applied to the development of fuzzy control systems with PI-fuzzy controllers dedicated to a class of integral type servo systems, where the linear case is focused on the IFT method in connection with the extended symmetrical optimum method to obtain the initial values of the linear PI controller parameters. Real-time experimental results corresponding to a fuzzy controlled nonlinear servo system are presented to validate the development method
{"title":"Genetic Iterative Feedback Tuning (GIFT) Method for Fuzzy Control System Development","authors":"R. Precup, S. Preitl","doi":"10.1109/ISEFS.2006.251133","DOIUrl":"https://doi.org/10.1109/ISEFS.2006.251133","url":null,"abstract":"This paper proposes an original iterative feedback tuning (IFT) method employing genetic algorithms to develop a class of fuzzy control systems. The approach is based on using the linear case results from the original IFT method and on replacing the parameter update law by genetic algorithms. Then, these results are transferred to the fuzzy case in terms of the modal equivalence principle resulting in an attractive development method referred to as genetic iterative feedback tuning (GIFT). The GIFT method is applied to the development of fuzzy control systems with PI-fuzzy controllers dedicated to a class of integral type servo systems, where the linear case is focused on the IFT method in connection with the extended symmetrical optimum method to obtain the initial values of the linear PI controller parameters. Real-time experimental results corresponding to a fuzzy controlled nonlinear servo system are presented to validate the development method","PeriodicalId":269492,"journal":{"name":"2006 International Symposium on Evolving Fuzzy Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2006-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129222766","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：