Pub Date : 2002-08-07DOI: 10.1109/NAFIPS.2002.1018102
E. Sanchez
Investigates a class of fuzzy relational equations, involving functional relations. It is shown how they can be solved, with ideas originating from the concept of truth-qualification of a fuzzy proposition. These equations are also discussed and related to the problem of the decomposition of a fuzzy relation by a fuzzy set.
{"title":"Functional relations and fuzzy relational equations","authors":"E. Sanchez","doi":"10.1109/NAFIPS.2002.1018102","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018102","url":null,"abstract":"Investigates a class of fuzzy relational equations, involving functional relations. It is shown how they can be solved, with ideas originating from the concept of truth-qualification of a fuzzy proposition. These equations are also discussed and related to the problem of the decomposition of a fuzzy relation by a fuzzy set.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"41 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":"132296416","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.2002.1018098
Ronald R. Yager, V. Kreinovich
In the arithmetic average, we combine all the estimates with equal weights. In some practical situations, it makes sense to give move weight to consistent estimates and less weight to estimates that axe far away from the consensus of the majority. Ordered weighted averaging (OWA) operators have been successfully applied in many practical problems. We explain this empirical success by showing that these operators are indeed guaranteed to work (i.e. universal), and that these operators are the best to use (in some reasonable sense).
{"title":"Main ideas behind OWA lead to a universal and optimal approximation scheme","authors":"Ronald R. Yager, V. Kreinovich","doi":"10.1109/NAFIPS.2002.1018098","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018098","url":null,"abstract":"In the arithmetic average, we combine all the estimates with equal weights. In some practical situations, it makes sense to give move weight to consistent estimates and less weight to estimates that axe far away from the consensus of the majority. Ordered weighted averaging (OWA) operators have been successfully applied in many practical problems. We explain this empirical success by showing that these operators are indeed guaranteed to work (i.e. universal), and that these operators are the best to use (in some reasonable sense).","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"336 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":"122643803","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.2002.1018054
D. Nauck
In this paper we describe ITEMS-a system for the estimation, visualization and exploration of travel data of a mobile workforce. One key feature of ITEMS is the interactive exploration of travel data that is visualized on maps. Users can not only see which journeys were late, on-time or early, but they can also request explanations why a journey was possibly late, for example. We have integrated a neuro-fuzzy system based on NEFCLASS into ITEMS. NEFCLASS generates explanatory fuzzy rules for a selected data subset in real time and presents them to the user. The rules can help the user in understanding the data better and in spotting possible problems in workforce management. We discuss aspects of learning interpretable fuzzy rules for generating explanations and demonstrate the application of NEFCLASS in the context of ITEMS.
{"title":"Neuro-fuzzy systems for explaining data sets","authors":"D. Nauck","doi":"10.1109/NAFIPS.2002.1018054","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018054","url":null,"abstract":"In this paper we describe ITEMS-a system for the estimation, visualization and exploration of travel data of a mobile workforce. One key feature of ITEMS is the interactive exploration of travel data that is visualized on maps. Users can not only see which journeys were late, on-time or early, but they can also request explanations why a journey was possibly late, for example. We have integrated a neuro-fuzzy system based on NEFCLASS into ITEMS. NEFCLASS generates explanatory fuzzy rules for a selected data subset in real time and presents them to the user. The rules can help the user in understanding the data better and in spotting possible problems in workforce management. We discuss aspects of learning interpretable fuzzy rules for generating explanations and demonstrate the application of NEFCLASS in the context of ITEMS.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"19 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":"126986946","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.2002.1018088
P. Bosc, O. Pivert
This paper deals with databases containing ill-known attribute values represented by possibility distributions. In order to manipulate such databases in a safe and efficient way, a constrained framework has been previously suggested, where a restricted number of querying operations are permitted (selections and foreign key joins calling on attributes taking imprecise values). The key for efficiency resides on the fact that these operators do not require to make computations explicitly over all the more or less possible worlds. An extension of this model involving the projection operator is proposed in this paper.
{"title":"About the projection operator in a possibilistic database framework","authors":"P. Bosc, O. Pivert","doi":"10.1109/NAFIPS.2002.1018088","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018088","url":null,"abstract":"This paper deals with databases containing ill-known attribute values represented by possibility distributions. In order to manipulate such databases in a safe and efficient way, a constrained framework has been previously suggested, where a restricted number of querying operations are permitted (selections and foreign key joins calling on attributes taking imprecise values). The key for efficiency resides on the fact that these operators do not require to make computations explicitly over all the more or less possible worlds. An extension of this model involving the projection operator is proposed in this paper.","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":"128863882","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.2002.1018047
J. C. Cano, P. Nava
Fuzzy systems rely on membership functions to represent input values for problem presentation and eventual problem solution. These can be generated in different ways, one of which is obtaining an expert to define the functions. This method is not always cost effective or available, so automatic membership function definition is extremely desirable Many methods for constructing membership functions based on knowledge engineering have been developed. Previous work has shown that statistical methods can be used to generate these membership functions. The quality of the result, however, is very application dependent. This study focuses on a method of automatic membership function generation that relies on the use of fuzzy relations. This paper describes the implementation of one such method, and examines its application to several data sets, including the identification of vowel sounds in spoken English.
{"title":"A fuzzy method for automatic generation of membership function using fuzzy relations from training examples","authors":"J. C. Cano, P. Nava","doi":"10.1109/NAFIPS.2002.1018047","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018047","url":null,"abstract":"Fuzzy systems rely on membership functions to represent input values for problem presentation and eventual problem solution. These can be generated in different ways, one of which is obtaining an expert to define the functions. This method is not always cost effective or available, so automatic membership function definition is extremely desirable Many methods for constructing membership functions based on knowledge engineering have been developed. Previous work has shown that statistical methods can be used to generate these membership functions. The quality of the result, however, is very application dependent. This study focuses on a method of automatic membership function generation that relies on the use of fuzzy relations. This paper describes the implementation of one such method, and examines its application to several data sets, including the identification of vowel sounds in spoken English.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"44 9 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":"124573967","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.2002.1018096
V. Kreinovich, H. Nguyen, S. Ferson, L. Ginzburg
Traditional interval computations provide an estimate for the result y=f(x/sub 1/,...,x/sub n/) of data processing when we know intervals x/sub 1/,...,x/sub n/ that are guaranteed to contain the (unknown) actual values of the quantities x/sub 1/,...,x/sub n/. Often, in addition to these guaranteed intervals, we have confidence intervals for these quantities, i.e., intervals x/sub i/ that contain the corresponding values x/sub i/ with a certain probability. It is desirable, based on the confidence intervals for x/sub i/, to produce the resulting confidence interval for y. It turns out that the formulas for computing such resulting confidence interval are closely related with the formulas for processing fuzzy numbers by using Zadeh's extension principle. Thus, known algorithms for processing fuzzy data can be used to process confidence intervals as well.
传统的区间计算提供了对结果y=f(x/下标1/，…)的估计。，x/下标n/)，当我们知道区间x/下标1/，…，x/下标n/，保证包含数量x/下标1/，…的(未知)实际值。x / an /。通常，除了这些保证区间之外，我们还有这些数量的置信区间，即区间x/下标i/以一定概率包含相应值x/下标i/。我们需要根据x/下标i/的置信区间来产生y的置信区间。结果表明，计算该置信区间的公式与利用Zadeh的可拓原理处理模糊数的公式密切相关。因此，已知的处理模糊数据的算法也可以用于处理置信区间。
{"title":"From computation with guaranteed intervals to computation with confidence intervals: a new application of fuzzy techniques","authors":"V. Kreinovich, H. Nguyen, S. Ferson, L. Ginzburg","doi":"10.1109/NAFIPS.2002.1018096","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018096","url":null,"abstract":"Traditional interval computations provide an estimate for the result y=f(x/sub 1/,...,x/sub n/) of data processing when we know intervals x/sub 1/,...,x/sub n/ that are guaranteed to contain the (unknown) actual values of the quantities x/sub 1/,...,x/sub n/. Often, in addition to these guaranteed intervals, we have confidence intervals for these quantities, i.e., intervals x/sub i/ that contain the corresponding values x/sub i/ with a certain probability. It is desirable, based on the confidence intervals for x/sub i/, to produce the resulting confidence interval for y. It turns out that the formulas for computing such resulting confidence interval are closely related with the formulas for processing fuzzy numbers by using Zadeh's extension principle. Thus, known algorithms for processing fuzzy data can be used to process confidence intervals as well.","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":"132486621","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.2002.1018094
S. Eschrich, L. Hall
Ensembles of classifiers often provide better classification accuracy than a single classifier. One approach to creating ensembles is to create different subsets of the training data. We present a method of creating ensembles of classifiers by partitioning the dataset into regions using clustering. Learners are assigned to each region and the ensemble classification occurs by querying the learned classifier. The first strategy considered for partitioning the training set is to generate a hard, non-overlapping partition. This approach is shown to perform worse than a single classifier using the entire training set. However, the use of soft partitions significantly improves the overall ensemble performance. Three different methods of creating soft partitions are considered: a simple distance ratio, and both the fuzzy c-means and possibilistic c-means membership functions. All three methods are found to improve overall classifier performance beyond hard partitioning and often perform better than the base classifier using the entire training set. Experiments on six datasets illustrate the improved accuracy from creating ensembles on soft partitions of data.
{"title":"Soft partitions lead to better learned ensembles","authors":"S. Eschrich, L. Hall","doi":"10.1109/NAFIPS.2002.1018094","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018094","url":null,"abstract":"Ensembles of classifiers often provide better classification accuracy than a single classifier. One approach to creating ensembles is to create different subsets of the training data. We present a method of creating ensembles of classifiers by partitioning the dataset into regions using clustering. Learners are assigned to each region and the ensemble classification occurs by querying the learned classifier. The first strategy considered for partitioning the training set is to generate a hard, non-overlapping partition. This approach is shown to perform worse than a single classifier using the entire training set. However, the use of soft partitions significantly improves the overall ensemble performance. Three different methods of creating soft partitions are considered: a simple distance ratio, and both the fuzzy c-means and possibilistic c-means membership functions. All three methods are found to improve overall classifier performance beyond hard partitioning and often perform better than the base classifier using the entire training set. Experiments on six datasets illustrate the improved accuracy from creating ensembles on soft partitions of data.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"2 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":"133626752","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.2002.1018038
Joao Paulo Carvalho, J. Tomé
This paper focuses on several stability issues regarding the modeling of the dynamics of qualitative real world systems, and the ability of fuzzy cognitive maps and rule-based fuzzy cognitive maps to provide a faithful modeling in what concerns the stability properties of those systems. It also introduces the concept of intrinsic stability as a necessary property of qualitative system dynamics modeling tools.
{"title":"Issues on the stability of fuzzy cognitive maps and rule-based fuzzy cognitive maps","authors":"Joao Paulo Carvalho, J. Tomé","doi":"10.1109/NAFIPS.2002.1018038","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018038","url":null,"abstract":"This paper focuses on several stability issues regarding the modeling of the dynamics of qualitative real world systems, and the ability of fuzzy cognitive maps and rule-based fuzzy cognitive maps to provide a faithful modeling in what concerns the stability properties of those systems. It also introduces the concept of intrinsic stability as a necessary property of qualitative system dynamics modeling tools.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"5 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":"131824316","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.2002.1018028
T. Whalen
Schweizer and Sklar (1961, 1963) define a family of T-norms from [0,1]X[0,1] to [0,1] using a parameter p as follows: T(a,b)=(/spl alpha//sup p/+ b/sup -p/-1)/sup -1/p/ if (a/sup -p/+b/sup -p/-1)/spl ges/0,0 otherwise. This paper considers the effects of removing the requirement that (a/sup -p/+b/sup -p/-1)/spl ges/0. This produces a family of complex improper T-norms, which can be used to define a corresponding family of real improper T-norms ranging from -1 to min(a,b). Improper strong implication functions created using the real improper T-norms support a variant of mode defuzzification, called best kernel defuzzification, with potentially useful properties for fuzzy expert systems.
{"title":"Improper strong implication","authors":"T. Whalen","doi":"10.1109/NAFIPS.2002.1018028","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018028","url":null,"abstract":"Schweizer and Sklar (1961, 1963) define a family of T-norms from [0,1]X[0,1] to [0,1] using a parameter p as follows: T(a,b)=(/spl alpha//sup p/+ b/sup -p/-1)/sup -1/p/ if (a/sup -p/+b/sup -p/-1)/spl ges/0,0 otherwise. This paper considers the effects of removing the requirement that (a/sup -p/+b/sup -p/-1)/spl ges/0. This produces a family of complex improper T-norms, which can be used to define a corresponding family of real improper T-norms ranging from -1 to min(a,b). Improper strong implication functions created using the real improper T-norms support a variant of mode defuzzification, called best kernel defuzzification, with potentially useful properties for fuzzy expert systems.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"3 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":"127323437","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.2002.1018048
U. Wagner
We present a methodology for semantic fuzzy sets. We construct alpha-cuts on the basis of observed data. Therefore we no longer need exclusively triangles, trapeziums or Gauss curves as elementary forms for fuzzy sets. In addition to that, we are able to integrate expert opinions, modelled as fuzzy sets. The methodology combines statistical interval estimation and distribution tests with fuzzy logic. It is applicable to random processes with an insufficient number of sample points. If the sample size increases, the result converges toward the statistical estimators. We applied the method to estimate the discharge of a river.
{"title":"Statistical based fuzzy sets","authors":"U. Wagner","doi":"10.1109/NAFIPS.2002.1018048","DOIUrl":"https://doi.org/10.1109/NAFIPS.2002.1018048","url":null,"abstract":"We present a methodology for semantic fuzzy sets. We construct alpha-cuts on the basis of observed data. Therefore we no longer need exclusively triangles, trapeziums or Gauss curves as elementary forms for fuzzy sets. In addition to that, we are able to integrate expert opinions, modelled as fuzzy sets. The methodology combines statistical interval estimation and distribution tests with fuzzy logic. It is applicable to random processes with an insufficient number of sample points. If the sample size increases, the result converges toward the statistical estimators. We applied the method to estimate the discharge of a river.","PeriodicalId":348314,"journal":{"name":"2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622)","volume":"70 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":"115763090","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：