Pub Date : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018049
M. Popescu, P. Gader
In this paper we develop an EM based training algorithm for a Takagi-Sugeno fuzzy rule system (TSFRS). Since the training is unsupervised, no target values are needed. The TSFRS models the degree of membership based on a given distribution that can be modified by changing the consequence of the rules or by rule pruning. We use this training algorithm to train a hidden Markov model (HMM) with state memberships provided by TSFRS using a modified Baum-Welch algorithm. This representation has the advantage of being transparent, since one can analyze and modify the rules that form the membership TSFRS.
{"title":"Continuous HMM with state memberships provided by Takagi-Sugeno fuzzy rule systems (TSFRS)","authors":"M. Popescu, P. Gader","doi":"10.1109/NAFIPS.2002.1018049","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018049","url":null,"abstract":"In this paper we develop an EM based training algorithm for a Takagi-Sugeno fuzzy rule system (TSFRS). Since the training is unsupervised, no target values are needed. The TSFRS models the degree of membership based on a given distribution that can be modified by changing the consequence of the rules or by rule pruning. We use this training algorithm to train a hidden Markov model (HMM) with state memberships provided by TSFRS using a modified Baum-Welch algorithm. This representation has the advantage of being transparent, since one can analyze and modify the rules that form the membership TSFRS.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"515 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116206551","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018119
A. Moussa, L. Kohout
Maintaining global network state information with unlimited precision in a dynamic large-scale network is practically impossible. The nature and the level of the inevitable imprecision vary significantly when different policies are used for updating the global network state information. This has a direct effect on the routing decisions and the network performance. We discuss the concept of precision of a network state, presenting a model for the impossibility of reaching, not only unlimited precision, but also the desired level of precision. We propose a novel network state updating policy to achieve two goals: bounded imprecision and much lower overhead by its ability to impede frequently unneeded updates without the need for a hold-down timer. The proposed policy is based on overlapping fuzzy tolerance classes and is partitioning-mechanism-independent. Hence, it can be used on top of currently available class-based policies.
{"title":"A new network state information updating policy using fuzzy tolerance classes and fuzzified precision","authors":"A. Moussa, L. Kohout","doi":"10.1109/NAFIPS.2002.1018119","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018119","url":null,"abstract":"Maintaining global network state information with unlimited precision in a dynamic large-scale network is practically impossible. The nature and the level of the inevitable imprecision vary significantly when different policies are used for updating the global network state information. This has a direct effect on the routing decisions and the network performance. We discuss the concept of precision of a network state, presenting a model for the impossibility of reaching, not only unlimited precision, but also the desired level of precision. We propose a novel network state updating policy to achieve two goals: bounded imprecision and much lower overhead by its ability to impede frequently unneeded updates without the need for a hold-down timer. The proposed policy is based on overlapping fuzzy tolerance classes and is partitioning-mechanism-independent. Hence, it can be used on top of currently available class-based policies.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120947704","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018077
S. Dick, C. Bethel, A. Kandel
We report on an experimental investigation of software reliability data. Our hypothesis in this investigation is that software failures are the result of a fundamentally deterministic process, rather than being realizations of a stochastic process as is commonly assumed. Using the techniques of nonlinear time series analysis, we examine three software reliability datasets for the signatures of deterministic, and possibly chaotic, behavior. In these datasets, we have found firm evidence of deterministic behavior, and hints of chaotic behavior. However, the latter are too limited to permit a definitive conclusion about the presence or absence of chaotic behavior.
{"title":"Are software failures chaotic?","authors":"S. Dick, C. Bethel, A. Kandel","doi":"10.1109/NAFIPS.2002.1018077","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018077","url":null,"abstract":"We report on an experimental investigation of software reliability data. Our hypothesis in this investigation is that software failures are the result of a fundamentally deterministic process, rather than being realizations of a stochastic process as is commonly assumed. Using the techniques of nonlinear time series analysis, we examine three software reliability datasets for the signatures of deterministic, and possibly chaotic, behavior. In these datasets, we have found firm evidence of deterministic behavior, and hints of chaotic behavior. However, the latter are too limited to permit a definitive conclusion about the presence or absence of chaotic behavior.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125612161","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018080
L. Kohout, Andreas Strotmann
We address the problem of collaboration between intelligent software systems with different competences across the Internet, focusing on some problems in the communication of knowledge between such systems. We point out the importance of exploiting the parallelism between knowledge communication amongst intelligent agents and that among people, exemplified by the compositionality principle and its importance in the engineering of knowledge communication languages, and by the computing-with-words paradigm and its importance in fuzzy systems. In this paper, we propose to exploit the parallelism in more detail, by employing the theory of categorial type logics (or more generally, of categorial grammars) in the design and implementation of symbolic knowledge communication languages and their fuzzy semantics for distributed knowledge-based systems.
{"title":"Effective computations with distributed knowledge: the issue of compositionality and ontologies","authors":"L. Kohout, Andreas Strotmann","doi":"10.1109/NAFIPS.2002.1018080","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018080","url":null,"abstract":"We address the problem of collaboration between intelligent software systems with different competences across the Internet, focusing on some problems in the communication of knowledge between such systems. We point out the importance of exploiting the parallelism between knowledge communication amongst intelligent agents and that among people, exemplified by the compositionality principle and its importance in the engineering of knowledge communication languages, and by the computing-with-words paradigm and its importance in fuzzy systems. In this paper, we propose to exploit the parallelism in more detail, by employing the theory of categorial type logics (or more generally, of categorial grammars) in the design and implementation of symbolic knowledge communication languages and their fuzzy semantics for distributed knowledge-based systems.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117307030","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018042
P. Blonda, A. Baraldi, A. D’Addabbo, C. Tarantino
In the field of image classification, this paper compares a traditional RBF two-stage hybrid learning procedure with an RBF two-stage learning technique exploiting labeled data to adapt hidden unit parameters. RBF centers are determined by running a clustering algorithm separately on different training sets, where each set is associated with a different class. The ELBG neural network is used as clustering algorithm. Two different data sets have been considered. The first consists of real three Synthetic Aperture Radar (SAR) image tandem pairs from ERS1/ERS2 satellites. The second consists of Magnetic Resonance (MR) slices of a patient affected by multiple sclerosis. The results indicate that the supervised approach performs better than the traditional approach when the number of hidden unit is the same and seems more stable to changes in the number of hidden units.
{"title":"RBF network with two-stage supervised learning: an application","authors":"P. Blonda, A. Baraldi, A. D’Addabbo, C. Tarantino","doi":"10.1109/NAFIPS.2002.1018042","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018042","url":null,"abstract":"In the field of image classification, this paper compares a traditional RBF two-stage hybrid learning procedure with an RBF two-stage learning technique exploiting labeled data to adapt hidden unit parameters. RBF centers are determined by running a clustering algorithm separately on different training sets, where each set is associated with a different class. The ELBG neural network is used as clustering algorithm. Two different data sets have been considered. The first consists of real three Synthetic Aperture Radar (SAR) image tandem pairs from ERS1/ERS2 satellites. The second consists of Magnetic Resonance (MR) slices of a patient affected by multiple sclerosis. The results indicate that the supervised approach performs better than the traditional approach when the number of hidden unit is the same and seems more stable to changes in the number of hidden units.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125347977","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018059
Young-Il Lee, C. Noe, Yong-Gi Kim
This paper describes the best choice of fuzzy implication operator and /spl alpha/-cut that are appropriate to the heuristic search technique for real-time collision avoidance of AUVs (autonomous underwater vehicles). A fuzzy implication operator is applied to the computation of a fuzzy triangle product that constructs a new fuzzy relation between two fuzzy relations. This is the theoretical basis of the heuristic search technique. In this paper, we design a heuristic search technique using fuzzy relational products for AUVs, and propose the selection of a fuzzy implication operator and /spl alpha/-cut which are the most suitable for the search technique. In order to verify the optimality and efficiency of the selected fuzzy implication operator and /spl alpha/-cut, we simulate every case of /spl alpha/-cut for each fuzzy implication operator in view of the cost of path and the number of /spl alpha/-cuts generating an acceptable path to the goal.
{"title":"Implication operators in fuzzy relational products for a local path-planning of AUVs","authors":"Young-Il Lee, C. Noe, Yong-Gi Kim","doi":"10.1109/NAFIPS.2002.1018059","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018059","url":null,"abstract":"This paper describes the best choice of fuzzy implication operator and /spl alpha/-cut that are appropriate to the heuristic search technique for real-time collision avoidance of AUVs (autonomous underwater vehicles). A fuzzy implication operator is applied to the computation of a fuzzy triangle product that constructs a new fuzzy relation between two fuzzy relations. This is the theoretical basis of the heuristic search technique. In this paper, we design a heuristic search technique using fuzzy relational products for AUVs, and propose the selection of a fuzzy implication operator and /spl alpha/-cut which are the most suitable for the search technique. In order to verify the optimality and efficiency of the selected fuzzy implication operator and /spl alpha/-cut, we simulate every case of /spl alpha/-cut for each fuzzy implication operator in view of the cost of path and the number of /spl alpha/-cuts generating an acceptable path to the goal.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130246082","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018095
C. Joslyn, J. C. Helton
We are interested in improving risk and reliability analysis of complex systems where our knowledge of system performance is provided by large simulation codes, and where moreover input parameters are known only imprecisely. Such imprecision lends itself to interval representations of parameter values, and thence to quantifying our uncertainty through Dempster-Shafer or Probability Bounds representations on the input space. In this context, the simulation code acts as a large "black box" function f, transforming one input Dempster-Shafer structure on the line into an output random interval f(A). Our quantification of output uncertainty is then based on this output random interval.. If some properties of f are known, then some information about f(A) can be determined. But when f is a pure black box, we must resort to sampling approaches. We present the basic formalism of a Monte Carlo approach to sampling a functionally propagated general random set, as opposed to a random interval. We show that the results of straightforward formal definitions are mathematically coherent, in the sense that bounding and convergence properties are achieved.
{"title":"Bounds on belief and plausibility of functionally propagated random sets","authors":"C. Joslyn, J. C. Helton","doi":"10.1109/NAFIPS.2002.1018095","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018095","url":null,"abstract":"We are interested in improving risk and reliability analysis of complex systems where our knowledge of system performance is provided by large simulation codes, and where moreover input parameters are known only imprecisely. Such imprecision lends itself to interval representations of parameter values, and thence to quantifying our uncertainty through Dempster-Shafer or Probability Bounds representations on the input space. In this context, the simulation code acts as a large \"black box\" function f, transforming one input Dempster-Shafer structure on the line into an output random interval f(A). Our quantification of output uncertainty is then based on this output random interval.. If some properties of f are known, then some information about f(A) can be determined. But when f is a pure black box, we must resort to sampling approaches. We present the basic formalism of a Monte Carlo approach to sampling a functionally propagated general random set, as opposed to a random interval. We show that the results of straightforward formal definitions are mathematically coherent, in the sense that bounding and convergence properties are achieved.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121212918","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018109
M. S. Chafi, M. Moavenian, M. Akbarzadeh-T.
The application of a novel fuzzy-neural architecture to diagnose faults in critical cases of a CNC X-axis drive system is described. The proposed architecture utilizes the concepts of fuzzy clustering, fuzzy decision making and RBF neural networks to create a suitable model based fault detection and isolation (FDI) structure. In the present application, the authors emphasize the faults due only to the nonlinear components and the components that have a more significant effect on overall accuracy of the drive system. On 100 tests on the system, i.e. the appropriate model, the diagnostic system allocated fault location and fault size 100 per cent correctly.
{"title":"Critical cases of a CNC drive system-fault diagnosis via a novel architecture","authors":"M. S. Chafi, M. Moavenian, M. Akbarzadeh-T.","doi":"10.1109/NAFIPS.2002.1018109","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018109","url":null,"abstract":"The application of a novel fuzzy-neural architecture to diagnose faults in critical cases of a CNC X-axis drive system is described. The proposed architecture utilizes the concepts of fuzzy clustering, fuzzy decision making and RBF neural networks to create a suitable model based fault detection and isolation (FDI) structure. In the present application, the authors emphasize the faults due only to the nonlinear components and the components that have a more significant effect on overall accuracy of the drive system. On 100 tests on the system, i.e. the appropriate model, the diagnostic system allocated fault location and fault size 100 per cent correctly.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125961624","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018099
T. Jobe, C. Helgason
We (2002) previously defined a necessary and sufficient causal ground to derive a formal causal ground in the unit hypercube. In addition we derived the clinical causal effect. These measures are developed from the fuzzy subset theorem as defined by Kosko (1992) and are in units of cardinality. In this paper we develop a measure of clinical efficiency K which is expressed in units of cardinality and is derived from the measures of formal causal ground and clinical causal effect. The measure of clinical efficiency was applied to 16 patients on an antiplatelet therapy. It renders a unique measure of clinical therapeutic effect for each patient. The measure of clinical efficiency developed herein can be used as an alternative to probability-based statistics to judge clinical causality.
{"title":"A fuzzy causal measure of individual patient treatment response","authors":"T. Jobe, C. Helgason","doi":"10.1109/NAFIPS.2002.1018099","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018099","url":null,"abstract":"We (2002) previously defined a necessary and sufficient causal ground to derive a formal causal ground in the unit hypercube. In addition we derived the clinical causal effect. These measures are developed from the fuzzy subset theorem as defined by Kosko (1992) and are in units of cardinality. In this paper we develop a measure of clinical efficiency K which is expressed in units of cardinality and is derived from the measures of formal causal ground and clinical causal effect. The measure of clinical efficiency was applied to 16 patients on an antiplatelet therapy. It renders a unique measure of clinical therapeutic effect for each patient. The measure of clinical efficiency developed herein can be used as an alternative to probability-based statistics to judge clinical causality.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121517813","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 : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018116
E. Walker
This paper surveys various aspects of image retrieval on the Internet. Many images are indexed by keyword or retrieved by similarity to a "key image". Existing work in areas such as linguistic variables for describing spatial relationships and color have natural applications in the area of image retrieval. Unlike image databases, the Internet is large and heterogeneous. The result of retrieval is necessarily a dialogue between the user (searcher) and the retrieval system. In this paper, several aspects of Internet information retrieval where fuzzy logic can be applied are highlighted.
{"title":"Image retrieval on the Internet-how can fuzzy help?","authors":"E. Walker","doi":"10.1109/NAFIPS.2002.1018116","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018116","url":null,"abstract":"This paper surveys various aspects of image retrieval on the Internet. Many images are indexed by keyword or retrieved by similarity to a \"key image\". Existing work in areas such as linguistic variables for describing spatial relationships and color have natural applications in the area of image retrieval. Unlike image databases, the Internet is large and heterogeneous. The result of retrieval is necessarily a dialogue between the user (searcher) and the retrieval system. In this paper, several aspects of Internet information retrieval where fuzzy logic can be applied are highlighted.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121528630","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：