Cumulative fuzzy class membership criterion decision-based classifier
Patrik Sabol, Peter Sinčák, Jan Buca, Pitoyo Hartono · 2017
This paper deals with classification algorithms as one of the basic principles of pattern recognition. We analyze their effect to a feature space and compare the type and the shape of the separating and decision surface, respectively. We proposed a novel classification approach based on Cumulative Fuzzy Membership Function that creates a decision surface in a different way as an MF ARTMAP neural network. We call the proposed decision surface Cumulative Fuzzy Class Membership Criterion (CFCMC), which we compared with the decision surface of MF ARTMAP termed as Membership Function. The analysis of both decision surfaces shows that CFCMC has better adaptability and flexibility in forming a decision boundary than Membership Function from MF ARTMAP classifier. Based on the result of this analysis we assumed that classifier built based on CFCMC should achieve higher classification accuracy than the one built based on Membership Function. Furthermore, we identified some issues, solutions and possible future challenges of our proposed novel method, such as the expansion into incremental learning and semantic information extraction.