By Topic

Feature extraction using Restricted Boltzmann Machine for stock price prediction

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Xianggao Cai ; School of Information Science and Technology, Sun Yat-sen University, Guangzhou, 510275, China ; Su Hu ; Xiaola Lin

Recently, many different types of artificial neural networks (ANNs) have been applied to forecast stock price and good performance is obtained. However, most of these models use only a small number of features as input and there may not be enough information to make prediction due to the complexity of stock market. If having a larger number of features, the run time of training would be increased and the generalization performance would be deteriorated due to the curse of dimension. Therefore, an effective tool to extract highly discriminative low-dimensional features from the high-dimensional raw input would be a great help in improving the generalization performance of the regression model. Restricted Boltzmann Machine (RBM) is a new type of machine learning tool with strong power of representation, which has been utilized as the feature extractor in a large variety of classification problems. In this paper, we use the RBM to extract discriminative low-dimensional features from raw data with dimension up to 324, and then use the extracted features as the input of Support Vector Machine (SVM) for regression. Experimental results indicate that our approach for stock price prediction has great improvement in terms of low forecasting errors compared with SVM using raw data.

Published in:

Computer Science and Automation Engineering (CSAE), 2012 IEEE International Conference on  (Volume:3 )

Date of Conference:

25-27 May 2012