Stock market time series data mining based on regularized neural network and rough set

Presents a method of stock market time series data mining, which combines a regularized neural network with rough sets. The process includes preprocessing of a time series database and data mining. The preprocessing cleans and filters time series. Then, we partition the time series into a series of static patterns, which is based on the trend (i.e., increasing or decreasing) of the closing price. An information table is formed by the most important predictable attributes and target attributes identified from each pattern. The regularized neural network (RNN) is used to study and predict the data. Rough sets can extract rule knowledge in the trained neural network that can be used to predict the time series behavior in the future. The method combines the high generalization faculty of the regularized neural network and the rule reduction capability of rough sets. An experiment demonstrates the effectiveness of the algorithm.