Knowledge-based data mining using quantitative and qualitative information on the internet

Abstract

Neural networks have shown considerable success in modeling financial data series. Neural networks have the ability to scan the data for patterns and can be used to construct nonlinear models. However, in financial data series forecasting, most neural network models are constructed on the basis of only quantitative information such as macroeconomic data. We divided the information, which affect the financial data series, into two categories, quantitative and qualitative factors. We proposed a method for integrating cognitive maps and neural networks to gain competitive advantage using qualitative information acquired from news information on the World Wide Web. We investigate ways to apply news information on the internet to the prediction of interest rates. We developed the KBNMiner (Knowledge-Based News Miner), which is designed to represent the knowledge of interest rate experts with cognitive maps (CMs), to search and retrieve news information on the internet according to prior knowledge, and to apply the information, which is retrieved from news information, to a neural network model for the prediction of interest rates. Our study focuses on improving the performance of data mining by using qualitative information. Real-world interest rate prediction data is used to illustrate the performance of the KBNMiner. Our integrated approach, which utilizes CMs and neural networks, has been shown to be effective in experiments. While the 10-fold cross validation is used to test our research model, the experimental results of the paired t-test have been found to be statistically significant. The 10-fold cross validation is used to test our research model and the experimental results of the paired t-test have been found to be statistically significant.