An Evaluation of an Algorithm for Inductive Learning of Bayesian Belief Networks Usin

TL;DR

This paper evaluates the K2 algorithm for inductive learning of Bayesian belief networks by testing it on simulated data and analyzing factors affecting its accuracy, also proposing a predictive model for its performance.

Contribution

It provides an empirical assessment of K2's accuracy and introduces a model to predict its performance based on data characteristics.

Findings

K2's accuracy varies with data set properties

A simple model can predict K2's performance

Simulation results highlight factors influencing learning success

Abstract

Bayesian learning of belief networks (BLN) is a method for automatically constructing belief networks (BNs) from data using search and Bayesian scoring techniques. K2 is a particular instantiation of the method that implements a greedy search strategy. To evaluate the accuracy of K2, we randomly generated a number of BNs and for each of those we simulated data sets. K2 was then used to induce the generating BNs from the simulated data. We examine the performance of the program, and the factors that influence it. We also present a simple BN model, developed from our results, which predicts the accuracy of K2, when given various characteristics of the data set.