This work improves on the FTNB algorithm to make it more tolerant of noise. The FTNB algorithm augments the Naïve Bayesian (NB) learning algorithm with a fine tuning stage in an attempt to find better estimations of the probability terms involved. The fine tuning stage has proved to be effective in improving the classification accuracy of the NB, however, it makes the NB algorithm more sensitive to noise in a training set.  This work presents several modifications of the fine turning stage in order to make it more tolerant of noise.

This work augments the Naïve Bayesian learning algorithm with a second training phase in an attempt to improve its classification accuracy. This is achieved by finding more accurate estimations of the needed probability terms. This approach helps in dealing with the problem of the lack of training data. Unlike many previous approaches that deal with this problem, the proposed method is an eager method in the sense that it does most of the work during training and, therefore, it does not increase classification time. It consists of two phases.

The classification accuracy of many machine learning methods depends upon their ability to accurately measure the similarity between different instances. Similarity is measured using a distance metric or measure. In this work, several novel distancemeasuresfornominalvaluesareproposed.Thesedistancemeasuresexploittheclassofatrainingexampleagainstwhich a new instance is compared. The experiments, conducted using 50 benchmark data sets, indicate that the proposed functions are superior in many cases to the Value Difference Metric (VDM) that is widely used in instance based learning.