Welcome to the AdPreqFr4SL web site

The AdPreqFr4SL learning framework for Bayesian Network Classifiers is designed to handle the cost-performance trade-off and cope with concept drift. Our strategy for incorporating new data is based on two main policies: bias management and gradual adaptation.

Starting with the simple Naive Bayes, we scale up the complexity by gradually updating attributes and structure. Since updating the structure is a costly task, we use new data to primarily adapt the parameters and only if this is really necessary, do we adapt the structure. The method for handling concept drift is based on the Shewhart P-Chart.

The method for handling concept drift is a simple, well-argued, statistically-driven method, as well as independent of the learning algorithm, which makes it broadly applicable

Main features

Cost-performance management

Simple control strategies based on the observation of some performance indicators to decide when to increase the k value and to start searching for new attribute dependences
This bias control leads to the selection of the optimal class-model for the current training data, avoiding the problems of underfitting or overfitting
Since updating the structure is a costly task, we reduce the cost of updating by first adapting parameters
The structure is adapted only at sparse time points, when there is some accumulated data and there is evidence that the use of the current structure no longer guarantees the desirable improvement in the performance

Concept drift handling

Based on Shewart's P-Chart (Statistical Quality Control)
A short-term memory is used to store the examples suspected to belong to a new concept
If a concept shift is detected, all the examples from the short-term memoty are used to build a new Naive Bayes classifier from scratch

Iterative Bayes for better probabability estimates

The iterative Bayes begins with the distribution tables built by the Naive Bayes algorithm.
Those tables are iteratively updated in order to improve the probability class distribution associated with each training example