Biologically-inspired evolving fuzzy ensemble

Neural fuzzy system is a hybrid intelligent system that synergizes artificial neural network and fuzzy logic techniques by combining the human-like reasoning style of fuzzy systems with the ability of learning and connected neural network structure. Existing neural networks have difficulty in learning from a dynamic data stream in the following aspects: 1) the limitation on the use of past explicit information poses difficulties in extending existing learning techniques; 2) the limited prior knowledge poses difficulties in designing a learning paradigm that can work without much prior information; 3) the sensitivity of learning methodology on the choice of the training parameters; 4) the unchangeable membership function width and 5) robustness of learning procedure against noisy or spurious data. In this project, novel biologically inspired neural-fuzzy computational models are presented to address the above mentioned difficulties encountered in the learning of a data stream. A newly computational model that integrates fuzzy system and neural network is proposed named eFE(evolving Fuzzy Ensemble). The eFE computational model employs an online learning algorithm that synthesizes error-driven learning theories and neurophysiological studies. The model has been implemented and run over several benchmarking simulations to demonstrate how accuracy and reliability the model has been achieved. Neural fuzzy systems have employed numerous applications in the financial field today. In this project, a proposed intelligent straddle trading system consists of volatility projection module(VPM) and trade decision module(TDM) will be implemented to help investor to make their trade decisions. The highly accurate eFE neural network provides the forecasting future stock price for the straddle trading system, so that by combining eFE and straddle trading system together, it will be more accurate and reliable for the trading decision made.Bachelor of Engineering (Computer Engineering