1 BC : a First - Order Bayesian Classi
نویسنده
چکیده
In this paper we present 1BC, a rst-order Bayesian Classier. Our approach is to view individuals as structured terms, and to distinguish between structural predicates referring to subterms (e.g. atoms from molecules), and properties applying to one or several of these subterms (e.g. a bond between two atoms). We describe an individual in terms of elementary features consisting of zero or more structural predicates and one property; these features are considered conditionally independent following the usual naive Bayes assumption. 1BC has been implemented in the context of the rst-order descriptive learner Tertius, and we describe several experiments demonstrating the viability of our approach.
منابع مشابه
1 BC : a First - Order Bayesian Classi
In this paper we present 1BC, a rst-order Bayesian Classi-er. Our approach is to view individuals as structured terms, and to distinguish between structural predicates referring to subterms (e.g. atoms from molecules), and properties applying to one or several of these sub-terms (e.g. a bond between two atoms). We describe an individual in terms of elementary features consisting of zero or more...
متن کاملBeyond Independence: Conditions for the Optimality of the Simple Bayesian Classi er
The simple Bayesian classi er (SBC) is commonly thought to assume that attributes are independent given the class, but this is apparently contradicted by the surprisingly good performance it exhibits in many domains that contain clear attribute dependences. No explanation for this has been proposed so far. In this paper we show that the SBC does not in fact assume attribute independence, and ca...
متن کاملBayesian Classi cation
Bayesian classi cation addresses the classi cation problem by learning the distribution of instances given di erent class values. We review the basic notion of Bayesian classi cation, describe in some detail the naive Bayesian classi er, and brie y discuss some extensions. C5.1.5.
متن کاملSearching for Dependencies in Bayesian Classifiers
Naive Bayesian classi ers which make independence assumptions perform remarkably well on some data sets but poorly on others. We explore ways to improve the Bayesian classi er by searching for dependencies among attributes. We propose and evaluate two algorithms for detecting dependencies among attributes and show that the backward sequential elimination and joining algorithm provides the most ...
متن کاملBeyond Independence: Conditions for the Optimality of the Simple Bayesian Classifier
The simple Bayesian classi er (SBC) is commonly thought to assume that attributes are independent given the class, but this is apparently contradicted by the surprisingly good performance it exhibits in many domains that contain clear attribute dependences. No explanation for this has been proposed so far. In this paper we show that the SBC does not in fact assume attribute independence, and ca...
متن کامل