Prediction of Short-Term Stock Price Trend Based on Multiview RBF Neural Network

Algorithm 1

MV-RBF algorithm.

	Step 1: Set the regularization parameters , as well as the RBF network's hidden layer nodes
Step 2: Use the FCM algorithm or other algorithm with the ability to divide the sample space to generate the data sets corresponding to the classification model of each view sample, and the corresponding cluster centers and width parameters ;
Step 3: Solve with formula (14) to get the Lagrange multiplier ，， ， ， ， ;
Step 4: According to the solution of dual problem formula (14) obtained in Step 3, use formulas (25a) and (25b) to find ;
Step 5: According to the parameters of each view obtained in Step 4, an RBF classification model that meets the characteristics of each view can be generated;