Pub Date : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851615
Juan Carlos Figueroa–García
This paper presents a representation of Type-2 fuzzy sets based on α-cuts and α-planes. Both representations can be used in different problems without loss of generality, so we show a complementary way to combine both of them in order to make computation of fuzzy functions easier. Some concluding remarks and recommendations are given for real applications.
{"title":"On α-representation of Type-2 fuzzy sets","authors":"Juan Carlos Figueroa–García","doi":"10.1109/NAFIPS.2016.7851615","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851615","url":null,"abstract":"This paper presents a representation of Type-2 fuzzy sets based on α-cuts and α-planes. Both representations can be used in different problems without loss of generality, so we show a complementary way to combine both of them in order to make computation of fuzzy functions easier. Some concluding remarks and recommendations are given for real applications.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122292739","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851607
Yingxu Wang
The fuzzy nature of language structures and semantics is formally studied towards a methodology for reasoning with fuzzy concepts (RFC). Mathematical models of fuzzy concepts and fuzzy semantics are introduced based on concept algebra and semantic algebra. The semantic effects of fuzzy modifiers and quantifiers on fuzzy concepts are quantitatively analyzed. Experiments on collective intension and extension elicitation for formal concepts demonstrate that fuzziness of human knowledge stem from the cognitive complexity, inexplicitness, subjectivity, diversity, redundancy, incompleteness, mixed synonyms, informal representation, incoherent attributes, divergent objects, and contextual influence. The RFC methodology provides a formal approach to computing with words (CW) for cognitive robots, deep machine learning, and fuzzy systems to rigorously manipulate fuzzy language entities, semantics, and reasoning in a wide range of applications.
{"title":"From computing with words (CWW) to reasoning with fuzzy concepts (RFC)","authors":"Yingxu Wang","doi":"10.1109/NAFIPS.2016.7851607","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851607","url":null,"abstract":"The fuzzy nature of language structures and semantics is formally studied towards a methodology for reasoning with fuzzy concepts (RFC). Mathematical models of fuzzy concepts and fuzzy semantics are introduced based on concept algebra and semantic algebra. The semantic effects of fuzzy modifiers and quantifiers on fuzzy concepts are quantitatively analyzed. Experiments on collective intension and extension elicitation for formal concepts demonstrate that fuzziness of human knowledge stem from the cognitive complexity, inexplicitness, subjectivity, diversity, redundancy, incompleteness, mixed synonyms, informal representation, incoherent attributes, divergent objects, and contextual influence. The RFC methodology provides a formal approach to computing with words (CW) for cognitive robots, deep machine learning, and fuzzy systems to rigorously manipulate fuzzy language entities, semantics, and reasoning in a wide range of applications.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115243206","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851577
Worrawate Leela-apiradee, P. Thipwiwatpotjana
Product pricing is one of the most important strategies in doing any business. In this paper, we propose a specific marketing situation as an acceptable product pricing problem when the data on purchasing power and transportation cost cannot be measured exactly but can be shown as intervals of possible values. This problem is formulated as a minimization problem of a differentiable convex objective function with max-plus interval linear constraints A ⊗ x = b where x is called an L-localized solution. Its feasible region is nonconvex but could be viewed by the union of box constraints. The steepest descent algorithm is applied to optimize the objective function with respect to each of these box constraints and obtained an optimal solution from the best value.
{"title":"Acceptable product pricing problem using L-localized solutions of max-plus interval linear equations","authors":"Worrawate Leela-apiradee, P. Thipwiwatpotjana","doi":"10.1109/NAFIPS.2016.7851577","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851577","url":null,"abstract":"Product pricing is one of the most important strategies in doing any business. In this paper, we propose a specific marketing situation as an acceptable product pricing problem when the data on purchasing power and transportation cost cannot be measured exactly but can be shown as intervals of possible values. This problem is formulated as a minimization problem of a differentiable convex objective function with max-plus interval linear constraints A ⊗ x = b where x is called an L-localized solution. Its feasible region is nonconvex but could be viewed by the union of box constraints. The steepest descent algorithm is applied to optimize the objective function with respect to each of these box constraints and obtained an optimal solution from the best value.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123500256","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851625
P. Melin, O. Castillo, Claudia I. González, J. R. Castro, O. Mendoza
Edge detection is an essential step used in image processing systems and can be applied to image sets before the training phase in pattern recognition systems to improve performance. An edge detector simplifies the analysis of the images; because, it reduces the data to be processed by highlighting the most important features. In this paper we show the advantage of using a fuzzy edge detector method in a face recognition system. In the proposed methodology, first the general type-2 fuzzy edge detector was applied over three image databases; secondly the recognition system was implemented using a monolithic neural network, and after that the mean recognition rate was obtained; finally the recognition rate is compared to other edge detectors, such as the Sobel operator, Type-1 and Interval Type-2 fuzzy edge detectors.
{"title":"General Type-2 fuzzy edge detectors applied to face recognition systems","authors":"P. Melin, O. Castillo, Claudia I. González, J. R. Castro, O. Mendoza","doi":"10.1109/NAFIPS.2016.7851625","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851625","url":null,"abstract":"Edge detection is an essential step used in image processing systems and can be applied to image sets before the training phase in pattern recognition systems to improve performance. An edge detector simplifies the analysis of the images; because, it reduces the data to be processed by highlighting the most important features. In this paper we show the advantage of using a fuzzy edge detector method in a face recognition system. In the proposed methodology, first the general type-2 fuzzy edge detector was applied over three image databases; secondly the recognition system was implemented using a monolithic neural network, and after that the mean recognition rate was obtained; finally the recognition rate is compared to other edge detectors, such as the Sobel operator, Type-1 and Interval Type-2 fuzzy edge detectors.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"2142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127467553","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851582
E. A. Newcomb, R. Hammell
Cyber defenders must make decisions under uncertainty using incomplete information. Information and communications networks are dynamic and complex; characteristics that contribute heavily to uncertainty.
{"title":"FLUF: Fuzzy logic utility framework to support computer network defense decision making","authors":"E. A. Newcomb, R. Hammell","doi":"10.1109/NAFIPS.2016.7851582","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851582","url":null,"abstract":"Cyber defenders must make decisions under uncertainty using incomplete information. Information and communications networks are dynamic and complex; characteristics that contribute heavily to uncertainty.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132070629","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851605
Christian F. Hempelmann, M. Petrenko, Gavin Matthews
This paper describes a method to automatically assign degrees of fuzzy set membership to individuals that have been asserted to be members of several classes. The method is tested in two variants on the case of persons who have several occupations as per Wikidata. While neither subclass is immediately successful, the new heuristic still proves to be sufficiently promising to document as an alternative to binary subclass assignment because it allows for degrees of membership that are emergent from existing knowledge bases rather than requiring manual assignment of arbitrary levels of crisp class membership.
{"title":"Automatic discovery of degrees of fuzzy set membership in ontologies","authors":"Christian F. Hempelmann, M. Petrenko, Gavin Matthews","doi":"10.1109/NAFIPS.2016.7851605","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851605","url":null,"abstract":"This paper describes a method to automatically assign degrees of fuzzy set membership to individuals that have been asserted to be members of several classes. The method is tested in two variants on the case of persons who have several occupations as per Wikidata. While neither subclass is immediately successful, the new heuristic still proves to be sufficiently promising to document as an alternative to binary subclass assignment because it allows for degrees of membership that are emergent from existing knowledge bases rather than requiring manual assignment of arbitrary levels of crisp class membership.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123612236","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851628
Gabriela E. Martinez, O. Mendoza, J. R. Castro, Antonio Rodríguez Díaz, P. Melin, O. Castillo
In this paper, a comparison of the Choquet and Sugeno integrals is presented. The proposed methods enable the calculation of the Choquet and Sugeno integrals to combine multiple source of information with a degree of uncertainty. The methods are used to combine the modules output of a modular neural network for face recognition and a comparison is performed. In this paper, the focus is on aggregation operators that use measures as inputs, in particular the Choquet and Sugeno integrals. Recognition results with the Choquet integral are better or comparable to results produced by Sugeno integral.
{"title":"Comparison between Choquet and Sugeno integrals as aggregation operators for pattern recognition","authors":"Gabriela E. Martinez, O. Mendoza, J. R. Castro, Antonio Rodríguez Díaz, P. Melin, O. Castillo","doi":"10.1109/NAFIPS.2016.7851628","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851628","url":null,"abstract":"In this paper, a comparison of the Choquet and Sugeno integrals is presented. The proposed methods enable the calculation of the Choquet and Sugeno integrals to combine multiple source of information with a degree of uncertainty. The methods are used to combine the modules output of a modular neural network for face recognition and a comparison is performed. In this paper, the focus is on aggregation operators that use measures as inputs, in particular the Choquet and Sugeno integrals. Recognition results with the Choquet integral are better or comparable to results produced by Sugeno integral.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128244968","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851597
Roberto Camacho Barranco, P. Teller
The accurate prediction of resource consumption is important when it comes to optimally scheduling jobs in heterogeneous computer systems, e.g., cloud and grid computing infrastructures. Accordingly, different methods have been proposed to estimate the computer resource consumption of applications executed on such systems. In this paper, we use neuro-fuzzy modeling to predict the resource consumption of two bioinformatics applications, RAxML and BLAST. We experiment with different numbers and shapes of the membership functions to obtain, from a broad test set, the best initial configuration, which is tuned using neuro-adaptive learning methods. The results obtained by the neuro-fuzzy models are compared with those of five differently configured machine-learning models using the Root Relative Squared Error of a ten-fold cross validation of each model. This comparison indicates that neuro-fuzzy modeling can be used to estimate computer resource consumption and can provide more accurate or competitively accurate predictions of execution-time consumption.
{"title":"Resource consumption prediction using neuro-fuzzy modeling","authors":"Roberto Camacho Barranco, P. Teller","doi":"10.1109/NAFIPS.2016.7851597","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851597","url":null,"abstract":"The accurate prediction of resource consumption is important when it comes to optimally scheduling jobs in heterogeneous computer systems, e.g., cloud and grid computing infrastructures. Accordingly, different methods have been proposed to estimate the computer resource consumption of applications executed on such systems. In this paper, we use neuro-fuzzy modeling to predict the resource consumption of two bioinformatics applications, RAxML and BLAST. We experiment with different numbers and shapes of the membership functions to obtain, from a broad test set, the best initial configuration, which is tuned using neuro-adaptive learning methods. The results obtained by the neuro-fuzzy models are compared with those of five differently configured machine-learning models using the Root Relative Squared Error of a ten-fold cross validation of each model. This comparison indicates that neuro-fuzzy modeling can be used to estimate computer resource consumption and can provide more accurate or competitively accurate predictions of execution-time consumption.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125403986","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851622
T. Tuan, Nguyen Hai Minh, Van Tao Nguyen, T. Ngan, To Huu Nguyen
In practical dentistry, dentists use their experience to examine dental X-ray images to identify patients symptoms for the diagnosis of possible diseases. However, this method is based solely on experts experience which varies from dentist to dentist. The idea of dental diagnosis from X-Ray images is to support dentists in making a more valid conclusion. In this paper, we propose a unified framework using Clustering and Fuzzy Rule-based systems for the diagnosis of dental problems. This framework is modeled under real dental cases of Hanoi Medical University, Vietnam including 56 dental images of 5 diseases in the period 2014 2015. Improvements of the standalone problems especially in the side of classification and decision making are demonstrated. Empirical results reveal the best method in terms of accuracy.
{"title":"Medical diagnosis from dental X-ray images: A novel approach using Clustering combined with Fuzzy Rule-based systems","authors":"T. Tuan, Nguyen Hai Minh, Van Tao Nguyen, T. Ngan, To Huu Nguyen","doi":"10.1109/NAFIPS.2016.7851622","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851622","url":null,"abstract":"In practical dentistry, dentists use their experience to examine dental X-ray images to identify patients symptoms for the diagnosis of possible diseases. However, this method is based solely on experts experience which varies from dentist to dentist. The idea of dental diagnosis from X-Ray images is to support dentists in making a more valid conclusion. In this paper, we propose a unified framework using Clustering and Fuzzy Rule-based systems for the diagnosis of dental problems. This framework is modeled under real dental cases of Hanoi Medical University, Vietnam including 56 dental images of 5 diseases in the period 2014 2015. Improvements of the standalone problems especially in the side of classification and decision making are demonstrated. Empirical results reveal the best method in terms of accuracy.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126497316","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 : 2016-10-01DOI: 10.1109/NAFIPS.2016.7851592
A. D. S. Farias, V. S. Costa, R. Santiago, B. Bedregal
A common problem found in literature is the reduction and minimization of automata. In fuzzy finite automata those processes are more complex; it is not always possible to minimize a given fuzzy automaton, M. In this paper we define a notion of order for the set of fuzzy Mealy machines type L-Valued and we prove the existence of a residuated function which works as operator of minimization of fuzzy Mealy machines.
{"title":"A residuated function in a class of Mealy type L-Valued finite automaton","authors":"A. D. S. Farias, V. S. Costa, R. Santiago, B. Bedregal","doi":"10.1109/NAFIPS.2016.7851592","DOIUrl":"https://doi.org/10.1109/NAFIPS.2016.7851592","url":null,"abstract":"A common problem found in literature is the reduction and minimization of automata. In fuzzy finite automata those processes are more complex; it is not always possible to minimize a given fuzzy automaton, M. In this paper we define a notion of order for the set of fuzzy Mealy machines type L-Valued and we prove the existence of a residuated function which works as operator of minimization of fuzzy Mealy machines.","PeriodicalId":208265,"journal":{"name":"2016 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132325653","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：