Fuzzy extreme learning machine for classification

Compared to traditional classifiers, such as SVM, the extreme learning machine (ELM) achieves similar performance for classification and runs at a much faster learning speed. However, in many real applications, the different input points may not be exactly assigned to one of the classes, such as the imbalance problems and the weighted classification problems. The traditional ELM lacks the ability to solve those problems. Proposed is a fuzzy ELM, which introduces a fuzzy membership to the traditional ELM method. Then, the inputs with different fuzzy matrix can make different contributions to the learning of the output weights. For the weighted classification problems, FELM can provide a more logical result than that of ELM.