The Integrated framework for the Prediction of Stock Price Index

Abstract

Integration of machine and human knowledge is more effective rather than a single kind of knowledge in solving unstructured problems. This paper proposes the integrated framework to achieve a better reasoning performance and comprehensive understanding on the stock price index prediction problem. Fuzzy neural network and the evaluation by expert respectively serve as the useful tools for generating the machine and human knowledge about the stock price index of next week. The machine and human knowledge are integrated by fuzzy logic-driven framework to generate the integrated knowledge. The conflicts among the integrated knowledge are solved by fuzzy rule base. The experimental results show that the proposed knowledge are solved by rule base. The experimental results show that the proposed knowledge integration significantly improves the reasoning performance.