Results 1 - 8 of 8

Fuzzy methods in machine learning and data mining: Status and prospects. Fuzzy Sets and Systems

by Eyke Hüllermeier , 2005
"... Over the past years, methods for the automated induction of models and the extraction of interesting patterns from empirical data have attracted considerable attention in the fuzzy set community. This paper briefly reviews some typical applications and highlights potential contributions that fuzzy s ..."
Abstract - Cited by 25 (0 self) - Add to MetaCart
Over the past years, methods for the automated induction of models and the extraction of interesting patterns from empirical data have attracted considerable attention in the fuzzy set community. This paper briefly reviews some typical applications and highlights potential contributions that fuzzy set theory can make to machine learning, data mining, and related fields. The paper concludes with a critical consideration of recent developments and some suggestions for future research directions. 1
(Show Context)

FR3: A fuzzy rule learner for inducing reliable classifiers

by Jens Hühn, Eyke Hüllermeier - IEEE Transactions Fuzzy Systems , 2009
"... This paper introduces a fuzzy rule-based classification method called FR3, which is short for ..."
Abstract - Cited by 12 (1 self) - Add to MetaCart
This paper introduces a fuzzy rule-based classification method called FR3, which is short for
(Show Context)

Fuzzy sets in machine learning and data mining

by Eyke Hüllermeier - Applied Soft Computing , 2008
"... Machine learning, data mining, and several related research areas are concerned with methods for the automated induction of models and the extraction of interesting patterns from empirical data. Automated knowledge acquisition of that kind has been an essential aspect of artificial intelligence sinc ..."
Abstract - Cited by 3 (0 self) - Add to MetaCart
Machine learning, data mining, and several related research areas are concerned with methods for the automated induction of models and the extraction of interesting patterns from empirical data. Automated knowledge acquisition of that kind has been an essential aspect of artificial intelligence since a long time and has more recently also attracted considerable attention in the fuzzy sets community. This paper briefly reviews some typical applications and highlights potential contributions that fuzzy set theory can make to machine learning, data mining, and related fields. In this connection, some advantages of fuzzy methods for representing and mining vague patterns in data are especially emphasized. 1
(Show Context)

Naive Possibilistic Classifiers for Imprecise or Uncertain Numerical Data

by Myriam Bounhas, Mohammad Ghasemi Hamed, Henri Prade, Serrurier Khaled Mellouli, Myriam Bounhas, Mohammad Ghasemi Hamed, Henri Prade, Mathieu Serrurier, Khaled Mel, Hal Id Hal, Myriam Bounhasa, Mohammad Ghasemi Hamedc, Henri Pradec, Serrurierc Khaled Melloulia , 2014
"... possibilistic classifiers for imprecise or uncertain ..."
Abstract - Add to MetaCart
possibilistic classifiers for imprecise or uncertain
(Show Context)

a r t i c l e i n f o Article history:

by unknown authors , 2012
"... research field for many researchers in pattern recognition and machine learning [4,100] and the study and development ngs to sup training se reference. Here, a pattern x follows the usual definition x x1; x2;...; xd;xf g, where d is the number of attributes t scribe the data and x is its assigned c ..."
Abstract - Add to MetaCart
research field for many researchers in pattern recognition and machine learning [4,100] and the study and development ngs to sup training se reference. Here, a pattern x follows the usual definition x x1; x2;...; xd;xf g, where d is the number of attributes t scribe the data and x is its assigned class. The general definition of the NN rule in supervised classification, the k nearest neighbors classifier (k-NN), consid use of the most similar (nearest) k patterns in TR to derive the class of a test pattern. More formally, let xi be a training
(Show Context)

Naive Possibilistic Classifiers for Imprecise or Uncertain Numerical Data

by Myriam Bounhasa, Mohammad Ghasemi Hamedc, Henri Pradec, Serrurierc Khaled Melloulia
"... In real-world problems, input data may be pervaded with uncertainty. In this paper, we investigate the behavior of naive possibilistic classifiers, as a counter-part to naive Bayesian ones, for dealing with classification tasks in presence of uncertainty. For this purpose, we extend possibilistic cl ..."
Abstract - Add to MetaCart
In real-world problems, input data may be pervaded with uncertainty. In this paper, we investigate the behavior of naive possibilistic classifiers, as a counter-part to naive Bayesian ones, for dealing with classification tasks in presence of uncertainty. For this purpose, we extend possibilistic classifiers, which have been recently adapted to numerical data, in order to cope with uncertainty in data representation. Here the possibility distributions that are used are supposed to encode the family of Gaussian probabilistic distributions that are compat-ible with the considered data set. We consider two types of uncertainty: i) the uncertainty associated with the class in the training set, which is modeled by a possibility distribution over class labels, and ii) the imprecision pervading attribute values in the testing set represented under the form of intervals for continuous data. Moreover, the approach takes into account the uncertainty about the estimation of the Gaussian distribution parameters due to the lim-ited amount of data available. We first adapt the possibilistic classification model, previously proposed for the certain case, in order to accommodate the uncertainty about class labels. Then, we propose an algorithm based on the extension principle to deal with imprecise attribute values. The experiments reported show the interest of possibilistic classifiers for handling uncertainty in data. In particular, the probability-to-possibility transform-based classifier shows a robust behavior when dealing with imperfect data.
(Show Context)

Dep.Telecommunications and Information Processing

by Parisa Kordjamshidi, Ghent University, Bernard De Baets, Guy De Tre
"... A fuzzy prototype-based method is introduced for learning from exam-ples. A special kind of prototype vector with fuzzy attributes is de-rived for each class from aggregat-ing fuzzified cases for the purpose of concept description. The fuzzi-fied cases are derived by defining a fuzzy membership func ..."
Abstract - Add to MetaCart
A fuzzy prototype-based method is introduced for learning from exam-ples. A special kind of prototype vector with fuzzy attributes is de-rived for each class from aggregat-ing fuzzified cases for the purpose of concept description. The fuzzi-fied cases are derived by defining a fuzzy membership function for each attribute of the sample cases. In a first method, for the classification of a new case, the membership degrees of its crisp attributes to fuzzy ag-gregated prototypes are measured. In a second method, after fuzzify-ing the new case, fuzzy set compar-ison methods are applied for mea-suring the similarity. The methods are compared to case-based ones like POSSIBL and kNN using UCI ma-chine learning repository. We also make comparisons by using various transformation methods from prob-abilities to possibilities instead of defining membership functions.
(Show Context)

Naive Possibilistic Classifiers for Imprecise or Uncertain Numerical Data

by Myriam Bounhas A, Mohammad Ghasemi Hamed C, Henri Prade C , 2014
"... In real-world problems, input data may be pervaded with uncertainty. In this paper, we investigate the behavior of naive possibilistic classifiers, as a counterpart to naive Bayesian ones, for dealing with classification tasks in presence of uncertainty. For this purpose, we extend possibilistic cla ..."
Abstract - Add to MetaCart
In real-world problems, input data may be pervaded with uncertainty. In this paper, we investigate the behavior of naive possibilistic classifiers, as a counterpart to naive Bayesian ones, for dealing with classification tasks in presence of uncertainty. For this purpose, we extend possibilistic classifiers, which have been recently adapted to numerical data, in order to cope with uncertainty in data representation. Here the possibility distributions that are used are supposed to encode the family of Gaussian probabilistic distributions that are compatible with the considered data set. We consider two types of uncertainty: i) the uncertainty associated with the class in the training set, which is modeled by a possibility distribution over class labels, and ii) the imprecision pervading attribute values in the testing set represented under the form of intervals for continuous data. Moreover, the approach takes into account the uncertainty about the estimation of the Gaussian distribution parameters due to the limited amount of data available. We first adapt the possibilistic classification
(Show Context)