Kononenko, I. (1990). "Comparison of Inductive and Naive Bayesian Learning Approaches to Automatic Knowledge Acquisition". Current Trends in Knowledge Adquisition, 190-197  Home/Search   Document Not in Database   Summary   Related Articles   Check

....assumption, a variety of empirical research shows surprisingly that the Naive Bayesian classifier can perform quite well compared to other methods even in domains where clear attribute dependencies exist. Furthermore, Naive Bayesian classifiers are also very simple and easy to understand (see [20]) In the Feature Decomposition approach with Naive Bayesian combination we use a similar idea, namely the prediction of a new instance x q is based on the product of the conditional probability of the target attribute, given the values of the input attributes in each subset. Mathematically it ....

Kononenko, I., "Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition". In Current Trends In reply to: Knowledge Acquisition, IOS Press, 1990.

....the document evaluated does this. When a new page arrives the system must predict how the user will evaluate it. To do that we compare the vector of features of this document with the user model for the Lecture Notes in Computer Science 9 current objective using an adaptation of the Naive Bayes [18] that has been proved to be a good classifier in [18] 28] 21] 16] 30] Objectives similarity. To find similar models is needed to compare different objectives. For this we use the Pearson Correlation [6] that we adapt to the representation of our synthesis model. In the eq. 1 w(a,i) is the ....

....arrives the system must predict how the user will evaluate it. To do that we compare the vector of features of this document with the user model for the Lecture Notes in Computer Science 9 current objective using an adaptation of the Naive Bayes [18] that has been proved to be a good classifier in [18] [28] 21] 16] 30] Objectives similarity. To find similar models is needed to compare different objectives. For this we use the Pearson Correlation [6] that we adapt to the representation of our synthesis model. In the eq. 1 w(a,i) is the similitude between the objectives a and i. v i,j is the ....

Kononenko, I. (1990). "Comparison of Inductive and Naive Bayesian Learning Approaches to Automatic Knowledge Acquisition". Current Trends in Knowledge Adquisition, 190-197

....estimated from the training examples, Naive Bayes is easy to construct. It is also, however, surprisingly effective. Many empirical comparisons between Naive Bayes and decision tree algorithms such as C4.5 (Quinlan, 1993) showed that Naive Bayes predicts just as well as C4.5 (Langley et al. 1992; Kononenko, 1990; Pazzani et al. 1996) However, the independence assumption hardly holds true in most artificial and real world datasets. For example, Frank et al. 2000) evaluated the performance of Naive Bayes on regression problems and compared it to numerical estimators on many artificial and realworld ....

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Ed.), Current trends in knowledge acquisition. IOS Press.

....and acceptably reliable and accurate. In practice, however, most combinations are not represented in the training data at all, let al..one in sufficient numbers to support accurate estimation of the 2 ZIJIAN ZHENG AND GEOFFREY I. WEBB required conditional probabilities. Naive Bayesian classification (Kononenko, 1990; Langley, Iba, Thompson, 1992; Langley Sage, 1994) circumvents this problem by assuming that all attributes are mutually independent within each class. This allows the following equality to be used: 2 P (V j C i ) Y v j 2V P (v j j C i ) 2) Naive Bayesian classifier learning is ....

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (eds.), Current Trends in Knowledge Acquisition. Amsterdam: IOS Press.

....order to show the generality of the framework, two different types of base learning algorithm are used in the following experiments. They are majority vote (MV) and the naive Bayesian classifier (NB) MV classifies all the test cases as belonging to the most common class of the training cases. NB [16, 17, 18] is an implementation of Bayes rule: P (C i jV ) P (C i )P (V jC i ) P (V ) for classification, where P denotes probability, C i is class i and V is a vector of attribute values describing a case. By assuming all attributes are mutually independent within each class, P (V jC i ) Q j P (v ....

Kononenko, I. Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (eds.), Current Trends in Knowledge Acquisition, 1990. Amsterdam: IOS Press.

....The voting scheme is simply summing up the vote for the predicted class of every individual model. 3 Naive Bayesian Classifier Learning and Levelled Naive Bayesian Tree Learning In this section, we describe two base learning algorithms. One is the naive Bayesian classifier (Duda Hart, 1973; Kononenko, 1990; Langley, Iba, Thompson, 1992) We will show that boosting does not significantly improve the performance of the naive Bayesian classifier. The other base learning algorithm is for generating levelled naive Bayesian trees. The objective of this algorithm is to investigate whether and how we can ....

Kononenko, I. (1990), Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (eds.), Current Trends in Knowledge Acquisition. Amsterdam: IOS Press.

....approach applies equally well to supervised learning tasks. Supervised Bayesian methods have long been used within the field of pattern recognition (Duda Hart, 1973) but only in the past few years have they received attention within the machine learning community (e.g. Clark Niblett, 1989; Kononenko, 1990, 1991; Langley, Iba, Thompson, 1992) In this paper we describe a technique designed to improve upon the already impressive behavior of the simplest approach to probabilistic induction the naive Bayesian classifier. Below we review the representational, performance, and learning assumptions ....

....represent a continuous probability distribution for each attribute. This requires that one assume some general form or model, with a common choice being the normal distribution, which can be conveniently represented entirely in terms of its mean and variance. 1. We have borrowed this term from Kononenko (1990); other common names for the method include the simple Bayesian classifier (Langley, 1993) and idiot Bayes (Buntine, 1990) Selective Bayesian Classifiers 400 To classify a new instance I, a naive Bayesian classifier applies Bayes theorem to determine the probability of each description given ....

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (Eds.), Current trends in knowledge acquisition. Amsterdam: IOS Press.

....probabilities P (C i ) and P (A k = V k j jC i ) may be estimated from the training data. To determine the most likely class of a test example, the probability of each class is computed with Equation 1. A classifier created in this manner is sometimes called a simple (Langley, 1993) or naive (Kononenko, 1990) Bayesian classifier. One important evaluation metric for machine learning methods is the predictive accuracy on unseen examples. This is measured by randomly selecting a subset of the examples in a database to use as training examples and reserving the remainder to be used as test examples. In ....

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Eds..), Current trends in knowledge acquisition. Amsterdam: IOS Press.

....accuracy of the naive Bayesian classifier on average. It performs better than the two approaches mentioned above in terms of higher prediction accuracy. 1 Introduction Naive Bayesian classifier learning is based on Bayes theorem and an attribute independence assumption (Duda and Hart 1973; Kononenko 1990; Langley and Sage 1994) Given training examples described using a vector of attribute values together with a known class for each example, the naive Bayesian classifier predicts the class of a new example V = v 1 ; v 2 ; Delta Delta Delta ; vn as the one with the highest probability of C ....

Kononenko, I.: Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (Eds.), Current Trends in Knowledge Acquisition. Amsterdam: IOS Press (1990).

....is worth ff k units. The voting scheme is simply summing up the vote for the predicted class of every individual model. 3 Naive Bayesian Classifier Learning and Levelled Naive Bayesian Tree Learning In this section, we describe two base learning algorithms. One is the naive Bayesian classifier [7, 11, 12]. We will show that boosting does not significantly improve the performance of the naive Bayesian classifier. The other base learning algorithm is for generating levelled naive Bayesian trees. The objective of this algorithm is to investigate whether and how we can make boosting perform better for ....

Kononenko, I.: Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (eds.), Current Trends in Knowledge Acquisition. Amsterdam: IOS Press. (1990).

No context found.

Kononenko, I. (1990). "Comparison of Inductive and Naive Bayesian Learning Approaches to Automatic Knowledge Acquisition". Current Trends in Knowledge Adquisition, 190-197

No context found.

Kononenko, I.: Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In Wielinga, B., ed.: Current Trends in Knowledge Acquisition. IOS Press (1990)

No context found.

I. Kononenko. Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga, editor, Current Trends in Knowledge Acquisition. IOS Press, 1990.

No context found.

I. Kononenko. Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga, J. Boose, B. Gaines, G. Schreiber, and M. van Someren, editors, Current Trends in Knowledge Acquisition. IOS Press, Amsterdam, 1990.

No context found.

Kononenko, I. (1990): Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In Wielinga, B. (Ed.), Current Trends in Knowledge Acquisition. IOS Press.

No context found.

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Eds..), Current trends in knowledge acquisition. Amsterdam: IOS Press.

No context found.

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Eds..), Current trends in knowledge acquisition. Amsterdam: IOS Press.

No context found.

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Ed.), Current Trends in Knowledge Acquisition. IOS Press, Amsterdam.

No context found.

Kononenko, I. (1990). Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga (Eds..), Current trends in knowledge acquisition. Amsterdam: IOS Press.

No context found.

Kaufmann. Kononenko, I. (1990) Comparison of inductive and naive Bayesian learning approaches to automatic knowledge acquisition. In B. Wielinga et al. (Eds.), Current Trends in Knowledge Acquisition. Amsterdam: IOS Press.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC