Evolving fuzzy Optimally Pruned Extreme Learning Machine: A comparative analysis

This paper proposes a method for the identification of evolving fuzzy Takagi-Sugeno systems based on the Optimally-Pruned Extreme Learning Machine (OP-ELM) methodology. We describe ELM which is a simple yet accurate and fast learning algorithm for training single-hidden layer feed-forward artificial neural networks (SLFNs) with random hidden neurons. We then describe the OP-ELM methodology for building ELM models in a robust and generic manner. Leveraging on the previously proposed Online Sequential ELM method and the OP-ELM, we propose an identification method for self-developing or evolving neuro-fuzzy systems. This method follows a random projection based approach to extracting evolving fuzzy rulebases. A comparison is performed over a diverse collection of datasets against well known evolving neuro-fuzzy methods, namely DENFIS and eTS. It is shown that the method proposed is robust and competitive in terms of accuracy and speed.