An experiment in integrating sentiment features for tech stock prediction in Twitter

Vu, Tien-Thanh; Chang, Shu; Ha, Quang Thuy and Collier, Nigel (2012). An experiment in integrating sentiment features for tech stock prediction in Twitter. In: Proceedings of the Workshop on Information Extraction and Entity Analytics on Social Media Data, The COLING 2012 Organizing Committee, Mumbai, India, pp. 23–38.

URL: http://www.aclweb.org/anthology/W12-5503

Abstract

Economic analysis indicates a relationship between consumer sentiment and stock price movements. In this study we harness features from Twitter messages to capture public mood related to four Tech companies for predicting the daily up and down price movements of these companies’ NASDAQ stocks. We propose a novel model combining features namely positive and negative sentiment, consumer confidence in the product with respect to ‘bullish’ or ‘bearish’ lexicon and three previous stock market movement days. The features are employed in a Decision Tree classifier using cross-fold validation to yield accuracies of 82.93%,80.49%, 75.61% and 75.00% in predicting the daily up and down changes of Apple (AAPL), Google (GOOG), Microsoft (MSFT) and Amazon (AMZN) stocks respectively in a 41 market day sample.

Viewing alternatives

Item Actions

Export

You can export this page using these formats