Argument Based Machine Learning in a Medical Domain

Argument Based Machine Learning (ABML) is a new approach to machine learning in which the learning examples could be accompanied by arguments. The arguments for specific examples are a special form of expert’s background knowledge which the expert uses to substantiate the class value for the chosen example. Možina et al. developed ABCN2 algorithm an extension of a well known rule learning algorithm CN2 that can use argumented examples in the learning process. In this work we present an application of ABCN2 in the medical domain which deals with severe bacterial infections in geriatric population. The elderly population, people over 65 years of age, is rapidly growing and it is estimated that it will double in the next 30 years. In Slovenia, it was accounted for over 15% in 2004 which is nearly the same as in USA and other developed countries. The costs of treating the patients aged over 65 are growing rapidly as well. In our study, we compare ABCN2 to CN2 and show that using arguments we improve the characteristics of the model. We also report the results that C4.5, Naïve Bayes and Logistic Regression achieve in this domain.