A Parameter Free BBN Discriminant Function for Optimum Model Complexity versus Goodness of Data Fitting

Abstract

Bayesian Belief Network (BBN) is an appealing classification model for learning causal and noncausal dependencies among a set of query variables. It is a challenging task to learning BBN structure from observational data because of pool of large number of candidate network structures. In this study, we have addressed the issue of goodness of data fitting versus model complexity. While doing so, we have proposed discriminant function which is non-parametric, free of implicit assumptions but delivering better classification accuracy in structure learning. The contribution in this study is twofold, first contribution (discriminant function) is in BBN structure learning and second contribution is for Decision Stump classifier. While designing the novel discriminant function, we analyzed the underlying relationship between the characteristics of data and accuracy of decision stump classifier. We introduced a meta characteristic measure AMfDS (herein known as Affinity Metric for Decision Stump) which is quite useful in prediction of classification accuracy of Decision Stump. AMfDS requires a single scan of the dataset.

An Empirical Analysis and Performance Evaluation of Feature Selection Techniques for Belief Network Classification System

“…It is already highlighted that feature selection can play an important role in evaluation of a classifier's performance. It is reported that little attention has been applied in evaluating the performance of BBN prior to its induction [11]. We can express our confidence that FSS is a key to estimation of performance of BBN prior to its induction phase in a real system.…”

“…They classified the techniques according to suitability, variety, usage and potential to sequence analysis and micorarray analysis. Although the pool of FSS techniques is becoming larger and larger [4,[10][11]; nevertheless specific exhaustive review leading to a wealth of comparative report for Bayesian belief network's various scoring function is not addressed so far. We in this study have incremented useful information in these survey reports; moreover our analysis is more precise in tweaking BBN in particular.…”

Mental Stress Classification Based on a Support Vector Machine and Naive Bayes Using Electrocardiogram Signals

“…This method defines the discriminant function using log likelihood. The log likelihood refers to a method of extracting parameters from a set of data while the discriminant function refers to a function that determines the category of a parameter [ 39 ]. The classification error function is defined by applying , which is the log likelihood of the feature vector , to the model parameter to Equation (12): …”