Results 1  10
of
25
Bayesian Network Classifiers
, 1997
"... Recent work in supervised learning has shown that a surprisingly simple Bayesian classifier with strong assumptions of independence among features, called naive Bayes, is competitive with stateoftheart classifiers such as C4.5. This fact raises the question of whether a classifier with less restr ..."
Abstract

Cited by 789 (23 self)
 Add to MetaCart
(Show Context)
Recent work in supervised learning has shown that a surprisingly simple Bayesian classifier with strong assumptions of independence among features, called naive Bayes, is competitive with stateoftheart classifiers such as C4.5. This fact raises the question of whether a classifier with less restrictive assumptions can perform even better. In this paper we evaluate approaches for inducing classifiers from data, based on the theory of learning Bayesian networks. These networks are factored representations of probability distributions that generalize the naive Bayesian classifier and explicitly represent statements about independence. Among these approaches we single out a method we call Tree Augmented Naive Bayes (TAN), which outperforms naive Bayes, yet at the same time maintains the computational simplicity (no search involved) and robustness that characterize naive Bayes. We experimentally tested these approaches, using problems from the University of California at Irvine repository, and compared them to C4.5, naive Bayes, and wrapper methods for feature selection.
Learning Bayesian network structure from massive datasets: the “sparse candidate” algorithm
 In Proceedings of the 15th Conference on Uncertainty in Artificial Intelligence (UAI
, 1999
"... Learning Bayesian networks is often cast as an optimization problem, where the computational task is to find a structure that maximizes a statistically motivated score. By and large, existing learning tools address this optimization problem using standard heuristic search techniques. Since the sear ..."
Abstract

Cited by 249 (9 self)
 Add to MetaCart
(Show Context)
Learning Bayesian networks is often cast as an optimization problem, where the computational task is to find a structure that maximizes a statistically motivated score. By and large, existing learning tools address this optimization problem using standard heuristic search techniques. Since the search space is extremely large, such search procedures can spend most of the time examining candidates that are extremely unreasonable. This problem becomes critical when we deal with data sets that are large either in the number of instances, or the number of attributes. In this paper, we introduce an algorithm that achieves faster learning by restricting the search space. This iterative algorithm restricts the parents of each variable to belong to a small subset of candidates. We then search for a network that satisfies these constraints. The learned network is then used for selecting better candidates for the next iteration. We evaluate this algorithm both on synthetic and reallife data. Our results show that it is significantly faster than alternative search procedures without loss of quality in the learned structures. 1
Learning Limited Dependence Bayesian Classifiers
 In KDD96: Proceedings of the Second International Conference on Knowledge Discovery and Data Mining
, 1996
"... We present a framework for characterizing Bayesian classification methods. This framework can be thought of as a spectrum of allowable dependence in a given probabilistic model with the Naive Bayes algorithm at the most restrictive end and the learning of full Bayesian networks at the most general e ..."
Abstract

Cited by 129 (4 self)
 Add to MetaCart
We present a framework for characterizing Bayesian classification methods. This framework can be thought of as a spectrum of allowable dependence in a given probabilistic model with the Naive Bayes algorithm at the most restrictive end and the learning of full Bayesian networks at the most general extreme. While much work has been carried out along the two ends of this spectrum, there has been surprising little done along the middle. We analyze the assumptions made as one moves along this spectrum and show the tradeoffs between model accuracy and learning speed which become critical to consider in a variety of data mining domains. We then present a general induction algorithm that allows for traversal of this spectrum depending on the available computational power for carrying out induction and show its application in a number of domains with different properties. Introduction Recently, work in Bayesian methods for classification has grown enormously (Cooper & Herskovits 1992) (Buntin...
Learning Goal Oriented Bayesian Networks for Telecommunications Risk Management
 In Proceedings of the 13th International Conference on Machine Learning
, 1996
"... This paper discusses issues related to Bayesian network model learning for unbalanced binary classification tasks. In general, the primary focus of current research on Bayesian network learning systems (e.g., K2 and its variants) is on the creation of the Bayesian network structure that fits the dat ..."
Abstract

Cited by 36 (0 self)
 Add to MetaCart
(Show Context)
This paper discusses issues related to Bayesian network model learning for unbalanced binary classification tasks. In general, the primary focus of current research on Bayesian network learning systems (e.g., K2 and its variants) is on the creation of the Bayesian network structure that fits the database best. It turns out that when applied with a specific purpose in mind, such as classification, the performance of these network models may be very poor. We demonstrate that Bayesian network models should be created to meet the specific goal or purpose intended for the model. We first present a goaloriented algorithm for constructing Bayesian networks for predicting uncollectibles in telecommunications riskmanagement datasets. Second, we argue and demonstrate that current Bayesian network learning methods may fail to perform satisfactorily in real life applications since they do not learn models tailored to a specific goal or purpose. Third, we discuss the performance of "goal oriented"...
Collective Mining of Bayesian Networks from Distributed Heterogeneous Data
, 2002
"... We present a collective approach to learning a Bayesian network from distributed heterogenous data. In this approach, we first learn a local Bayesian network at each site using the local data. Then each site identifies the observations that are most likely to be evidence of coupling between local an ..."
Abstract

Cited by 24 (7 self)
 Add to MetaCart
(Show Context)
We present a collective approach to learning a Bayesian network from distributed heterogenous data. In this approach, we first learn a local Bayesian network at each site using the local data. Then each site identifies the observations that are most likely to be evidence of coupling between local and nonlocal variables and transmits a subset of these observations to a central site. Another Bayesian network is learnt at the central site using the data transmitted from the local site. The local and central Bayesian networks are combined to obtain a collective Bayesian network, that models the entire data. Experimental results and theoretical justification that demonstrate the feasibility of our approach are presented.
Probabilistic conflicts in a search algorithm for estimating posterior probabilities in Bayesian networks
, 1996
"... This paper presents a search algorithm for estimating posterior probabilities in discrete Bayesian networks. It shows how conflicts (as used in consistencybased diagnosis) can be adapted to speed up the search. This algorithm is especially suited to the case where there are skewed distributions, al ..."
Abstract

Cited by 23 (6 self)
 Add to MetaCart
This paper presents a search algorithm for estimating posterior probabilities in discrete Bayesian networks. It shows how conflicts (as used in consistencybased diagnosis) can be adapted to speed up the search. This algorithm is especially suited to the case where there are skewed distributions, although nothing about the algorithm or the definitions depends on skewness of distributions. The general idea is to forward simulate the network, based on the `normal' values for each variable (the value with high probability given its parents). When a predicted value is at odds with the observations, we analyse which variables were responsible for the expectation failure  these form a conflict  and continue forward simulation considering different values for these variables. This results in a set of possible worlds from which posterior probabilities  together with error bounds  can be 1 derived. Empirical results with Bayesian networks having tens of thousands of nodes are presented.
Independency relationships and learning algorithms for singly connected networks
, 1998
"... Graphical structures such as Bayesian networks or M arkov networks are very useful tools for representing irrelevance or independency relationships, and they may be used to efficiently perform reasoning tasks. Singly connected networks are important specific cases where there is no more than one un ..."
Abstract

Cited by 19 (10 self)
 Add to MetaCart
Graphical structures such as Bayesian networks or M arkov networks are very useful tools for representing irrelevance or independency relationships, and they may be used to efficiently perform reasoning tasks. Singly connected networks are important specific cases where there is no more than one undirected path connecting each pair of variables. The aim of this paper is to investigate the kind of properties that a dependency model must verify in order to be equivalent to a singly connected graph structure, as a way of driving automated discovery and construction of singly connected networks in data. The main results are the characterizations of those dependency models which are isomorphic to singly connected graphs (either via the dseparation criterion for directed acyclic graphs or via the separation criterion for undirected graphs), as well as the development of efficient algorithms for learning singly connected graph representations of dependency models.
Knowledge discovery in telecommunication services data using Bayesian Models
 In Proceedings of the First International Conference on Knowledge Discovery (KDD95
, 1993
"... Fraud and uncollectible debt are multibillion dollar problems in the telecommunications industry. Because it is difficult to know which accounts will go bad, we are faced with the difficult knowledgediscovery task of characterizing a rare binary outcome using large amounts of noisy, highdimension ..."
Abstract

Cited by 18 (1 self)
 Add to MetaCart
(Show Context)
Fraud and uncollectible debt are multibillion dollar problems in the telecommunications industry. Because it is difficult to know which accounts will go bad, we are faced with the difficult knowledgediscovery task of characterizing a rare binary outcome using large amounts of noisy, highdimensional data. Binary characterizatioos may be of interest but will not be especially useful in this domain. Instead, proposing an action requires an estimate of the probability that a customer or a call is uncollectible. This paper addresses the discovery of predictive knowledge bearing on fraud and uncollectible debt using a supervised machine leamiog method that constructs Bayesiao network models. The new method is able to predict rare event outcomes and cope with the quirks and copious amounts of input data. The Bay&an network models it produces serve as ao input module to a normative decisionsupport system and suggest ways to reinforce or redirect existing efforts in the problem area. We compare the performance of several conditionally independent models with the conditionally dependent models discovered by the new learning system using realworld datasets of 46 million records and 603 800 million bytes. I.
Designing food with Bayesian belief networks
 in I. Parmee, ed., `Evolutionary Design and Manufacture { ACDM2000
, 2000
"... Abstract. The food industry is highly competitive, and in order to survive, manufacturers must constantly innovate and match the ever changing tastes of consumers. A recent survey [1] found that 90 % of the 13,000 new food products launched each year in the US fail within one year. Food companies ar ..."
Abstract

Cited by 9 (1 self)
 Add to MetaCart
(Show Context)
Abstract. The food industry is highly competitive, and in order to survive, manufacturers must constantly innovate and match the ever changing tastes of consumers. A recent survey [1] found that 90 % of the 13,000 new food products launched each year in the US fail within one year. Food companies are therefore changing the way new products are developed and launched, and this includes the use of intelligent computer systems. This paper provides an overview of one particular technique, namely Bayesian Belief networks, and its application to a typical food design problem. The characteristics of an &quot;ideal &quot; product are derived from a small data set. 1.
Interactive Genetic Algorithm Framework for Long Term Groundwater Monitoring Design
 Proceedings of the Environmental & Water Resources Institute (EWRI) World Water & Environmental Resources Congress, ASCE, Salt Lake City, UT
, 2004
"... In standard optimization approaches for water resources management problems, the designer is responsible for correctly formulating mathematical equations to describe the system objectives and constraints. The search for optimal or nearoptimal solutions is made under the assumption that these formul ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
In standard optimization approaches for water resources management problems, the designer is responsible for correctly formulating mathematical equations to describe the system objectives and constraints. The search for optimal or nearoptimal solutions is made under the assumption that these formulated objectives and constraints completely describe the system. However, in real systems that is often not true. Many qualitative criteria can be integral parts of the design analysis that numerically based algorithms cannot capture. For such problems, designer interaction with the search algorithm can help the search be more creative and inclusive. Genetic algorithms are ideally suited for incorporating such interaction in their usual search process, and can successfully evolve solutions that are optimal with respect to both qualitative and quantitative objectives. Under an interactive approach, the genetic algorithm performs the usual operations of selection, crossover, and mutation, but the user evaluates the suitability (‘fitness’) of candidate solutions, enabling objectives that cannot be quantified to be included in the search process. In multiobjective problems, where quantitative objectives can be as important as qualitative fitness of designs, analysis of designs is done based on tradeoff fronts made from both quantitative and qualitative information. In this paper, we demonstrate the use of interactive genetic algorithms for long term groundwater monitoring problems, which have multiple numerical and subjective objectives. We also analyze the effects on the optimal monitoring designs of using an interactive optimization approach instead of more traditional numerical optimization approaches.