Combining machine learning and metaheuristics algorithms for classification method PROAFTN

Abstract

The supervised learning classification algorithms are one of the most well known successful techniques for ambient assisted living environments. However the usual supervised learning classification approaches face issues that limit their application especially in dealing with the knowledge interpretation and with very large unbalanced labeled data set. To address these issues fuzzy classification method PROAFTN was proposed. PROAFTN is part of learning algorithms and enables to determine the fuzzy resemblance measures by generalizing the concordance and discordance indexes used in outranking methods. The main goal of this chapter is to show how the combined meta-heuristics with inductive learning techniques can improve performances of the PROAFTN classifier. The improved PROAFTN classifier is described and compared to well known classifiers, in terms of their learning methodology and classification accuracy. Through this chapter we have shown the ability of the metaheuristics when embedded to PROAFTN method to solve efficiency the classification problems.