Weighted Fuzzy Similarity Classifier in the Łukasiewicz-Structure

The aim of this paper is to introduce improvements made to a classifier based on the fuzzy similarity (1). Improvements are based on the use of generalized Łukasiewicz-structure and weight optimization. We are presenting some new results and a more detailed description of the theoretical background and fixing some terminology compared in to our previous work (2). The main benefits of the classifier are its computational efficiency and its strong mathematical background. It is based on many-valued logic and it provides semantic information about classification results. We will show that if we choose the power value in appropriate manner in the generalized Łukasiewicz-structure and the optimal weights for different features, we will see significant enhancements in classification results.