Pub Date : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494438
Cleo Pau, Temur Kutsia
We consider the problem of solving approximate equations between logic terms. The approximation is expressed by proximity relations. They are reflexive and symmetric (but not necessarily transitive) fuzzy binary relations. The equations are solved by variable substitutions that bring the sides of equations “close” to each other with respect to a predefined degree. We consider unification and matching equations in which mismatches in function symbol names, arity, and in the argument order are tolerated (i.e., the approximate equations are formulated over so called fully fuzzy signatures). This work generalizes on the one hand, class-based proximity unification to fully fuzzy signatures, and on the other hand, unification with similarity relations over a fully fuzzy signature by extending similarity to proximity.
{"title":"Proximity-Based Unification and Matching for Fully Fuzzy Signatures","authors":"Cleo Pau, Temur Kutsia","doi":"10.1109/FUZZ45933.2021.9494438","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494438","url":null,"abstract":"We consider the problem of solving approximate equations between logic terms. The approximation is expressed by proximity relations. They are reflexive and symmetric (but not necessarily transitive) fuzzy binary relations. The equations are solved by variable substitutions that bring the sides of equations “close” to each other with respect to a predefined degree. We consider unification and matching equations in which mismatches in function symbol names, arity, and in the argument order are tolerated (i.e., the approximate equations are formulated over so called fully fuzzy signatures). This work generalizes on the one hand, class-based proximity unification to fully fuzzy signatures, and on the other hand, unification with similarity relations over a fully fuzzy signature by extending similarity to proximity.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128770377","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494523
S. Zadrożny, J. Kacprzyk, Mateusz Dziedzic
We propose a new approach to database querying, termed context seeking querying, which involves context that is crucial for information interpretation and understanding yet practically not considered in querying. We present a justification, formalization and two algorithms for the new queries.
{"title":"A Concept of Context-Seeking Queries","authors":"S. Zadrożny, J. Kacprzyk, Mateusz Dziedzic","doi":"10.1109/FUZZ45933.2021.9494523","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494523","url":null,"abstract":"We propose a new approach to database querying, termed context seeking querying, which involves context that is crucial for information interpretation and understanding yet practically not considered in querying. We present a justification, formalization and two algorithms for the new queries.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129053749","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494521
Penugonda Ravikumar, R. U. Kiran, N. Unnam, Y. Watanobe, K. Goda, V. Devi, P. K. Reddy
Time series classification is an important model in data mining. It involves assigning a class label to a test instance based on the training data with known class labels. Most previous studies developed time series classifiers by disregarding the fuzzy nature of events (i.e., events with similar values may belong to different classes) within the data. Consequently, these studies suffered from performance issues, including decreased accuracy and increased memory, runtime, and energy requirements. With this motivation, this paper proposes a novel fuzzy nearest neighbor classifier for time series data. The basic idea of our classifier is to transform the very large training data into a relatively small representative training data and use it to label a test instance by employing a new fuzzy distance measure known as Ravi. Experimental results on real world benchmark datasets demonstrate that the proposed classifier outperforms the current parameter-free time series classifiers and also the popular deep learning techniques.
{"title":"A Novel Parameter-Free Energy Efficient Fuzzy Nearest Neighbor Classifier for Time Series Data","authors":"Penugonda Ravikumar, R. U. Kiran, N. Unnam, Y. Watanobe, K. Goda, V. Devi, P. K. Reddy","doi":"10.1109/FUZZ45933.2021.9494521","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494521","url":null,"abstract":"Time series classification is an important model in data mining. It involves assigning a class label to a test instance based on the training data with known class labels. Most previous studies developed time series classifiers by disregarding the fuzzy nature of events (i.e., events with similar values may belong to different classes) within the data. Consequently, these studies suffered from performance issues, including decreased accuracy and increased memory, runtime, and energy requirements. With this motivation, this paper proposes a novel fuzzy nearest neighbor classifier for time series data. The basic idea of our classifier is to transform the very large training data into a relatively small representative training data and use it to label a test instance by employing a new fuzzy distance measure known as Ravi. Experimental results on real world benchmark datasets demonstrate that the proposed classifier outperforms the current parameter-free time series classifiers and also the popular deep learning techniques.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127833541","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494345
H. Sadłowska, A. Kochański, P. Grzegorzewski
Hydroforming is a relatively new technology of forming and profiling. So far, the application of this method has been limited by the costs of die production. The cost of the dies and the long production start-up time made this method economically viable for the production of hundreds of products. The approach change to the tool design for profile shaping techniques has allowed to develop the new hydroforming method perfectly suited to low-volume or even unit production. In traditional solutions, the die is rigid and does not deform during the expansion of the profile. In the newly patented RTH (Rapid Tube Hydroforming) method, the die undergoes controlled deformation during the process. The specificity of the granular materials used for the production of the dies makes modeling the behavior of the die during the expansion of the profile a remarkable problem. This contribution presents considerations on the fuzzy inference method used to model the technological process. As a result, it was possible to more accurately determine the importance of individual die parameters (geometry and material properties), and thus better predict the final shape of the formed profile. The main goal is to understand the effect of shaped profile on the matrix and to recognize the influence of granular material in the matrix under the compaction conditions of the expanded profile on its final geometry.
{"title":"Multi-Phase Fuzzy Modeling in the Innovative RTH Hydroforming Technology","authors":"H. Sadłowska, A. Kochański, P. Grzegorzewski","doi":"10.1109/FUZZ45933.2021.9494345","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494345","url":null,"abstract":"Hydroforming is a relatively new technology of forming and profiling. So far, the application of this method has been limited by the costs of die production. The cost of the dies and the long production start-up time made this method economically viable for the production of hundreds of products. The approach change to the tool design for profile shaping techniques has allowed to develop the new hydroforming method perfectly suited to low-volume or even unit production. In traditional solutions, the die is rigid and does not deform during the expansion of the profile. In the newly patented RTH (Rapid Tube Hydroforming) method, the die undergoes controlled deformation during the process. The specificity of the granular materials used for the production of the dies makes modeling the behavior of the die during the expansion of the profile a remarkable problem. This contribution presents considerations on the fuzzy inference method used to model the technological process. As a result, it was possible to more accurately determine the importance of individual die parameters (geometry and material properties), and thus better predict the final shape of the formed profile. The main goal is to understand the effect of shaped profile on the matrix and to recognize the influence of granular material in the matrix under the compaction conditions of the expanded profile on its final geometry.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"350 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120941342","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494404
M. Ojeda-Hernández, I. P. Cabrera, P. Cordero, Emilio Muñoz-Velasco
Two alternative definitions of closure system in complete fuzzy lattices are introduced, first as a crisp set and then as a fuzzy one. It is valuated in a complete Heyting algebra and follows the classical definition on complete lattices. The classical bijection between closure systems and fuzzy closure operators is preserved. Then, the notion is compared with the most used definition given by Bělohlávek on the fuzzy powerset lattice.
{"title":"On (fuzzy) closure systems in complete fuzzy lattices","authors":"M. Ojeda-Hernández, I. P. Cabrera, P. Cordero, Emilio Muñoz-Velasco","doi":"10.1109/FUZZ45933.2021.9494404","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494404","url":null,"abstract":"Two alternative definitions of closure system in complete fuzzy lattices are introduced, first as a crisp set and then as a fuzzy one. It is valuated in a complete Heyting algebra and follows the classical definition on complete lattices. The classical bijection between closure systems and fuzzy closure operators is preserved. Then, the notion is compared with the most used definition given by Bělohlávek on the fuzzy powerset lattice.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120962887","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494566
G. Gallo, M. Bernardi, Marta Cimitile, P. Ducange
The increasing number of always more sophisticated car sensors, which allow to extract information about the driver, encourages auto vehicle developers and researchers to focus on the topic of driver identification. The advantages can be various, such as to customise and improve driver experience, to increase safety and to reduce global environmental problems. This work explores a set of features extracted from a car monitoring system, installed on real cars, to identify the driver on the basis of his/her driving behaviour. The proposed features are leveraged by a Multiobjective Evolutionary Learning Scheme for generating Fuzzy Rule-Based Classifiers characterized by different trade-offs between the classification accuracy and the explainability of the classification models. To evaluate the effectiveness and efficiency of the proposed approach, we carry out an experimental analysis on a real-world dataset, composed by actual measures extracted from 4 cars driven by 4 different drivers. The results show that the fuzzy classification models experimented in this work are more accurate and explaninable than the classification models generated adopting tree-based classifiers, such as decision trees and random forests.
{"title":"An Explainable Approach for Car Driver Identification","authors":"G. Gallo, M. Bernardi, Marta Cimitile, P. Ducange","doi":"10.1109/FUZZ45933.2021.9494566","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494566","url":null,"abstract":"The increasing number of always more sophisticated car sensors, which allow to extract information about the driver, encourages auto vehicle developers and researchers to focus on the topic of driver identification. The advantages can be various, such as to customise and improve driver experience, to increase safety and to reduce global environmental problems. This work explores a set of features extracted from a car monitoring system, installed on real cars, to identify the driver on the basis of his/her driving behaviour. The proposed features are leveraged by a Multiobjective Evolutionary Learning Scheme for generating Fuzzy Rule-Based Classifiers characterized by different trade-offs between the classification accuracy and the explainability of the classification models. To evaluate the effectiveness and efficiency of the proposed approach, we carry out an experimental analysis on a real-world dataset, composed by actual measures extracted from 4 cars driven by 4 different drivers. The results show that the fuzzy classification models experimented in this work are more accurate and explaninable than the classification models generated adopting tree-based classifiers, such as decision trees and random forests.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116328692","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494577
Beatriz Laiate, R. Watanabe, E. Esmi, F. S. Pedro, L. C. Barros
Based on the concept of cross product, defined for fuzzy numbers in general, this paper introduces an operation of multiplication for special classes of fuzzy numbers. Each one of these classes of fuzzy numbers is a vector space generated by strongly independent fuzzy numbers isomorphic to $mathbb{R}_^n$. It proves that under some conditions, these vector spaces of fuzzy numbers are closed under this operation of multiplication. In addition, some algebraic properties of this operation are listed. Lastly, a notion of fuzzy division as an inverse operation of the product is provided.
{"title":"A cross product of $mathcal{S}$-linearly correlated fuzzy numbers","authors":"Beatriz Laiate, R. Watanabe, E. Esmi, F. S. Pedro, L. C. Barros","doi":"10.1109/FUZZ45933.2021.9494577","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494577","url":null,"abstract":"Based on the concept of cross product, defined for fuzzy numbers in general, this paper introduces an operation of multiplication for special classes of fuzzy numbers. Each one of these classes of fuzzy numbers is a vector space generated by strongly independent fuzzy numbers isomorphic to $mathbb{R}_^n$. It proves that under some conditions, these vector spaces of fuzzy numbers are closed under this operation of multiplication. In addition, some algebraic properties of this operation are listed. Lastly, a notion of fuzzy division as an inverse operation of the product is provided.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126370457","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494431
M. Deardorff, Derek T. Anderson, T. Havens, Bryce J. Murray, S. Kakula, Timothy Wilkin
This paper focuses on a powerful nonlinear aggregation function, the Choquet integral (ChI). Specifically, we focus on situations where the parameters of the ChI are learned from data. For N inputs, the ChI breaks down into N! underlying linear convex sums (LCSs) with 2N shared variables. Typically, these LCSs are reducible into a drastically smaller number of linear order statistics (LOSs). In the spirit of explainable AI (XAI), our goal is to discover the minimal underlying operator structure of a learned ChI to be conveyed to its users. The challenge is, there does not appear to be widespread research or agreement regarding how to compute similarity within and between measures or integrals. In this paper, we explore the earth mover's distance (EMD), a parametric cross-bin measure, to capture semantic relatedness between LOSs. EMD is used to measure dissimilarity between integrals. In the case of a single ChI, underlying aggregation operator structure is discovered via EMD and clustering. A combination of synthetic and real-world experiments are provided to demonstrate interpretability and reduction of complexity.
{"title":"Earth Mover's Distance as a Similarity Measure for Linear Order Statistics and Fuzzy Integrals","authors":"M. Deardorff, Derek T. Anderson, T. Havens, Bryce J. Murray, S. Kakula, Timothy Wilkin","doi":"10.1109/FUZZ45933.2021.9494431","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494431","url":null,"abstract":"This paper focuses on a powerful nonlinear aggregation function, the Choquet integral (ChI). Specifically, we focus on situations where the parameters of the ChI are learned from data. For N inputs, the ChI breaks down into N! underlying linear convex sums (LCSs) with 2N shared variables. Typically, these LCSs are reducible into a drastically smaller number of linear order statistics (LOSs). In the spirit of explainable AI (XAI), our goal is to discover the minimal underlying operator structure of a learned ChI to be conveyed to its users. The challenge is, there does not appear to be widespread research or agreement regarding how to compute similarity within and between measures or integrals. In this paper, we explore the earth mover's distance (EMD), a parametric cross-bin measure, to capture semantic relatedness between LOSs. EMD is used to measure dissimilarity between integrals. In the case of a single ChI, underlying aggregation operator structure is discovered via EMD and clustering. A combination of synthetic and real-world experiments are provided to demonstrate interpretability and reduction of complexity.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127400776","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494420
M. Skublewska-Paszkowska, Paweł Karczmarek, E. Lukasik
In tennis there are two basic shots (forehand and backhand), which are two of key elements to win points. Sophisticated equipments, such as motion capture systems, enable one to record both the tennis player's movements and tennis racket. The 3D data may be used to define the perfect shot model or to give the directions to the player how to reach to this model and which aspects of impact should be improved. Clustering analysis can result in understanding the phases of a tennis player move and, as a consequence, the improvement of his/her play. Using the memberships obtained in the fuzzy clustering process one can evaluate the quality of a player's move and potentially estimate the player's progress. The main objective of this study is to apply the fuzzy c-means algorithm utilizing the dynamic time warping-based distance to cluster analysis of tennis shots. Both shots were taken into the consideration. The analysis consists of forty moves. Based on the 3D data of the tennis racket positions, the clustering was performed for subsequent two, three, and four clusters. The obtained results clearly show that clustering with two clusters is the most appropriate for analysing tennis shots. The model of a perfect shot was obtained. It is universal and does not depend on the player's height. Based on the model, it is possible to deduce technical differences in the players' shots. This analysis gives the directions for improvements of the shot technique. The advantage of the clustering of our approach is that we can get information to what degree the athlete should still correct his/her shots. The information is given to what extent the stroke is correct in relation to the ideal model.
{"title":"Tennis Multivariate Time Series Clustering","authors":"M. Skublewska-Paszkowska, Paweł Karczmarek, E. Lukasik","doi":"10.1109/FUZZ45933.2021.9494420","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494420","url":null,"abstract":"In tennis there are two basic shots (forehand and backhand), which are two of key elements to win points. Sophisticated equipments, such as motion capture systems, enable one to record both the tennis player's movements and tennis racket. The 3D data may be used to define the perfect shot model or to give the directions to the player how to reach to this model and which aspects of impact should be improved. Clustering analysis can result in understanding the phases of a tennis player move and, as a consequence, the improvement of his/her play. Using the memberships obtained in the fuzzy clustering process one can evaluate the quality of a player's move and potentially estimate the player's progress. The main objective of this study is to apply the fuzzy c-means algorithm utilizing the dynamic time warping-based distance to cluster analysis of tennis shots. Both shots were taken into the consideration. The analysis consists of forty moves. Based on the 3D data of the tennis racket positions, the clustering was performed for subsequent two, three, and four clusters. The obtained results clearly show that clustering with two clusters is the most appropriate for analysing tennis shots. The model of a perfect shot was obtained. It is universal and does not depend on the player's height. Based on the model, it is possible to deduce technical differences in the players' shots. This analysis gives the directions for improvements of the shot technique. The advantage of the clustering of our approach is that we can get information to what degree the athlete should still correct his/her shots. The information is given to what extent the stroke is correct in relation to the ideal model.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127243238","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 : 2021-07-11DOI: 10.1109/FUZZ45933.2021.9494569
Ravikiran Chimatapu, H. Hagras, M. Kern, G. Owusu
The recent advances in the field of Artificial Intelligence (AI) have led to the rapid deployment of AI systems in a variety of fields such as healthcare, financial, education etc. However, many of the AI systems are black boxes which restricts the use of these AI in applications that are highly regulated (such as financial, justice, medical, autonomous vehicles etc.) where it is necessary to provide satisfactory explanations for the decisions taken. A variety of approaches that have been proposed to tackle this problem, but these approaches generally emphasize providing satisfactory explanations for individual predictions at the cost of providing explanations at the global level. Hence, to solve these problems, in this paper, we present a hybrid deep learning type-2 fuzzy logic system which addresses these challenges by providing a highly interpretable model that can be trained using both labelled and unlabeled data. We also present a method to extract global and local explanations for this model. We also show that the presented model has reasonable performance when compared to stacked autoencoders deep neural networks.
{"title":"Enhanced Deep Type-2 Fuzzy Logic System For Global Interpretability","authors":"Ravikiran Chimatapu, H. Hagras, M. Kern, G. Owusu","doi":"10.1109/FUZZ45933.2021.9494569","DOIUrl":"https://doi.org/10.1109/FUZZ45933.2021.9494569","url":null,"abstract":"The recent advances in the field of Artificial Intelligence (AI) have led to the rapid deployment of AI systems in a variety of fields such as healthcare, financial, education etc. However, many of the AI systems are black boxes which restricts the use of these AI in applications that are highly regulated (such as financial, justice, medical, autonomous vehicles etc.) where it is necessary to provide satisfactory explanations for the decisions taken. A variety of approaches that have been proposed to tackle this problem, but these approaches generally emphasize providing satisfactory explanations for individual predictions at the cost of providing explanations at the global level. Hence, to solve these problems, in this paper, we present a hybrid deep learning type-2 fuzzy logic system which addresses these challenges by providing a highly interpretable model that can be trained using both labelled and unlabeled data. We also present a method to extract global and local explanations for this model. We also show that the presented model has reasonable performance when compared to stacked autoencoders deep neural networks.","PeriodicalId":151289,"journal":{"name":"2021 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)","volume":"221 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127289413","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：