Hybrid intelligence approaches for designing a dynamic financial time-series predictive model based on web-architecture home finance learning environment

This study proposes design concepts for a comprehensive home financial learning environment that individual investors can use as a reference in establishing web-based learning and investment platforms. This study also introduces a hybrid approach that demonstrates a data mining function of the financial learning environment. Known as Fuzzy BPN, this approach is comprised of backpropagation neural network (BPN) and fuzzy membership function. This membership function takes advantage of the nonlinear features of artificial neural networks (ANNs) and the interval values as a means of overcoming the inadequacy of single-point estimation of ANNs. Based from these characteristics, a dynamic and intelligent time-series forecasting system will be developed for practical financial predictions. In addition to this, the experimental processing can demonstrate the feasibility of applying the hybrid model-Fuzzy BPN. The empirical results of the study show that Fuzzy BPN provides an alternative data mining tool for financial learning environment to investment forecasting.