#### DMCA

## A Naive Bayes Style Possibilistic Classifier

### Cached

### Download Links

Citations: | 8 - 0 self |

### Citations

8903 |
Probabilistic reasoning in intelligent systems: networks of plausible inference
- Pearl
- 1988
(Show Context)
Citation Context ... object and can be seen as a special type of probabilistic networks. Probabilistic networks, which have been studied extensively in the field of graphical modeling [11] and of which Bayesian networks =-=[19]-=- and Markov networks [15] are the most popular, are based on the idea to exploit independence relations between attributes used to describe a domain in order to decompose a multivariate probability di... |

4843 |
Pattern Classification and Scene Analysis
- Duda, Hart
- 1973
(Show Context)
Citation Context ...a naive Bayes classifier is a special probabilistic network. Thus we obtain a classifier that can easily handle imprecise information in the data to learn from. 1 Introduction Naive Bayes classifiers =-=[8, 5, 13]-=- are an old and wellknown type of classifiers. They use a probabilistic approach to assign a class to a case or an object and can be seen as a special type of probabilistic networks. Probabilistic net... |

1524 |
Local computations with probabilities on graphical structures and their applications to expert systems (with discussion),
- Lauritzen, Spiegelhalter
- 1988
(Show Context)
Citation Context ...s a special type of probabilistic networks. Probabilistic networks, which have been studied extensively in the field of graphical modeling [11] and of which Bayesian networks [19] and Markov networks =-=[15]-=- are the most popular, are based on the idea to exploit independence relations between attributes used to describe a domain in order to decompose a multivariate probability distribution into a set of ... |

1397 | A Bayesian method for the induction of probabilistic networks from data
- Cooper, Herskovits
- 1992
(Show Context)
Citation Context ...pecific decomposition of a given multivariate probability distribution. Naive Bayes classifiers as well as the more general probabilistic networks can easily be induced from a dataset of sample cases =-=[14, 4, 10]-=-. A drawback of all probabilistic approaches is that they provide a framework to handle uncertain, but precise information. However, the explicit treatment of imprecise (i.e., set-valued) information ... |

1157 | Learning Bayesian networks: The combination of knowledge and statistical data.
- Heckerman, Geiger, et al.
- 1995
(Show Context)
Citation Context ...pecific decomposition of a given multivariate probability distribution. Naive Bayes classifiers as well as the more general probabilistic networks can easily be induced from a dataset of sample cases =-=[14, 4, 10]-=-. A drawback of all probabilistic approaches is that they provide a framework to handle uncertain, but precise information. However, the explicit treatment of imprecise (i.e., set-valued) information ... |

855 |
UCI Repository of machine learning databases.
- Murphy, Aha
- 1994
(Show Context)
Citation Context ...7 Experimental Results We implemented the suggested possibilistic classifier along with a normal naive Bayes classifier and tested both on four datasets from the UC Irvine machine learning repository =-=[16]-=-. In both cases we used the simplification procedures described in the preceding section. The results are shown in table 1, together with the results obtained with a decision tree classifier. The colu... |

439 | An analysis of Bayesian classifier
- Langley, Iba, et al.
- 1992
(Show Context)
Citation Context ...a naive Bayes classifier is a special probabilistic network. Thus we obtain a classifier that can easily handle imprecise information in the data to learn from. 1 Introduction Naive Bayes classifiers =-=[8, 5, 13]-=- are an old and wellknown type of classifiers. They use a probabilistic approach to assign a class to a case or an object and can be seen as a special type of probabilistic networks. Probabilistic net... |

265 | Induction of selective bayesian classifiers,
- Langley, Sage
- 1994
(Show Context)
Citation Context ...pecific decomposition of a given multivariate probability distribution. Naive Bayes classifiers as well as the more general probabilistic networks can easily be induced from a dataset of sample cases =-=[14, 4, 10]-=-. A drawback of all probabilistic approaches is that they provide a framework to handle uncertain, but precise information. However, the explicit treatment of imprecise (i.e., set-valued) information ... |

199 |
The Estimation of Probabilities: An Essay on Modern Bayesian Methods.
- Good
- 1965
(Show Context)
Citation Context ...a naive Bayes classifier is a special probabilistic network. Thus we obtain a classifier that can easily handle imprecise information in the data to learn from. 1 Introduction Naive Bayes classifiers =-=[8, 5, 13]-=- are an old and wellknown type of classifiers. They use a probabilistic approach to assign a class to a case or an object and can be seen as a special type of probabilistic networks. Probabilistic net... |

179 |
Hugin - a shell for building bayesian belief universes for expert systems.
- Andersen, Olesen, et al.
- 1989
(Show Context)
Citation Context ...omain in order to decompose a multivariate probability distribution into a set of (conditional or marginal) distributions on lower dimensional subspaces. Early efficient implementations include HUGIN =-=[1]-=- and PATHFINDER [9]. Naive Bayes classifiers are a special case of probabilistic networks, since they make strong assumptions about the dependence and independence relations between the class attribut... |

159 |
Foundations of Fuzzy Systems
- Kruse, Gebhardt, et al.
- 1994
(Show Context)
Citation Context ..., which due to their connection to fuzzy systems and their ability to deal not only with uncertainty but also with imprecision, recently gained some attention. They have been implemented in POSSINFER =-=[7, 12]-=-. In this paper, we focus on the latter approach, that is, on possibilistic networks. We suggest a classifier that is a special case of a possibilistic network in much the same way in which a naive Ba... |

150 | Probabilistic Similarity Networks,
- Heckerman
- 1990
(Show Context)
Citation Context ...ecompose a multivariate probability distribution into a set of (conditional or marginal) distributions on lower dimensional subspaces. Early efficient implementations include HUGIN [1] and PATHFINDER =-=[9]-=-. Naive Bayes classifiers are a special case of probabilistic networks, since they make strong assumptions about the dependence and independence relations between the class attribute and the other att... |

74 |
Valuation-based systems: A framework for managing uncertainty in expert systems
- Shenoy
- 1991
(Show Context)
Citation Context ...reatment of imprecise information usually renders the corresponding inference mechanisms computationally intractable. A noteworthy attempt in this direction are the so-called valuation-based networks =-=[21, 22]-=- which have been implemented in PULCINELLA [20]. Another are possibilistic networks, which due to their connection to fuzzy systems and their ability to deal not only with uncertainty but also with im... |

62 | ANeuro-Fuzzy Method to learn Fuzzy Classification Rules from Data', - Nauck, Kruse - 1997 |

50 | PULCINELLA: a general tool for propagating uncertainty in valuation networks
- SaÆotti, Umkehrer
- 1991
(Show Context)
Citation Context ...he corresponding inference mechanisms computationally intractable. A noteworthy attempt in this direction are the so-called valuation-based networks [21, 22] which have been implemented in PULCINELLA =-=[20]-=-. Another are possibilistic networks, which due to their connection to fuzzy systems and their ability to deal not only with uncertainty but also with imprecision, recently gained some attention. They... |

34 |
An introduction to random sets.
- Nguyen
- 2006
(Show Context)
Citation Context ...In this model possibility distributions are seen as informationcompressed representations of (not necessarily nested) random sets and a degree of possibility as the one-point coverage of a random set =-=[18]-=-. To be more precise: Let ! 0 be the actual, but unknown state of a domain of interest, which is contained in a set\Omega of possible states. Let (T ; 2 T ; P ), T = ft 1 ; t 2 ; : : : ; t r g, be a f... |

26 |
The Context Model — An Integrating View of Vagueness and Uncertainty Int. Journal of Approximate Reasoning 9:283–314
- Gebhardt, Kruse
- 1993
(Show Context)
Citation Context ...ybe the best way to explain the difference between uncertain and imprecise information is to consider the notion of a degree of possibility. The interpretation we prefer is based on the context model =-=[6, 12]-=-. In this model possibility distributions are seen as informationcompressed representations of (not necessarily nested) random sets and a degree of possibility as the one-point coverage of a random se... |

19 |
Local Computations in Hypertrees (Working Paper 201
- Shafer, Shenoy
- 1988
(Show Context)
Citation Context ...reatment of imprecise information usually renders the corresponding inference mechanisms computationally intractable. A noteworthy attempt in this direction are the so-called valuation-based networks =-=[21, 22]-=- which have been implemented in PULCINELLA [20]. Another are possibilistic networks, which due to their connection to fuzzy systems and their ability to deal not only with uncertainty but also with im... |

16 |
POSSINFER — A Software Tool for Possibilistic Inference
- Gebhardt, Kruse
- 1996
(Show Context)
Citation Context ..., which due to their connection to fuzzy systems and their ability to deal not only with uncertainty but also with imprecision, recently gained some attention. They have been implemented in POSSINFER =-=[7, 12]-=-. In this paper, we focus on the latter approach, that is, on possibilistic networks. We suggest a classifier that is a special case of a possibilistic network in much the same way in which a naive Ba... |

14 |
Uncertainty and Vagueness
- Kruse, Schwecke, et al.
- 1991
(Show Context)
Citation Context ...ch to assign a class to a case or an object and can be seen as a special type of probabilistic networks. Probabilistic networks, which have been studied extensively in the field of graphical modeling =-=[11]-=- and of which Bayesian networks [19] and Markov networks [15] are the most popular, are based on the idea to exploit independence relations between attributes used to describe a domain in order to dec... |

11 | Efficient maximum projection of database-induced multivariate possibility distributions
- Borgelt, Kruse
- 1998
(Show Context)
Citation Context ...ut poses some problems of efficiency. In contrast to conditional probabilities, which can easily be computed from case counters (see above), maximum projections are much harder to obtain. However, in =-=[3]-=- we suggested a solution to this problem. This solution has proven to be efficient for many applications (although pathological cases can be constructed). It is based on dataset num. of possibilistic ... |

5 |
Chapter F1.2: Inference Methods
- Borgelt, Gebhardt, et al.
- 1998
(Show Context)
Citation Context ...obabilistic approaches can handle uncertain, but precise information. As a concept of independence, which is fundamental to the technique of graphical modeling, we use possibilistic non-interactivity =-=[2]-=-. Let X , Y , and Z be three disjoint subsets of attributes. Then X is called conditionally independent of Y given Z w.r.t. , if 8! 2\Omega : (!X[Y j !Z ) = minf(!X j !Z ); (! Y j !Z )g whenever (!Z )... |