Impact of Online Stock Reviews on Stock Market Trends Using Text Sentiment Analysis Methods

Authors

  • Haixiang Li School of Economics and Management, Xi’an University of Technology, Xi’an 710048, Shaanxi, China
  • Weixian Xue School of Humanities and Management, Xi’an Traffic Engineering Institute, Xi’an 710300, Shaanxi, China

Abstract

As an important part of the financial market, the stock exchange market plays a very important role in the development of the economy. To better predict stock trends, this article proposes a text sentiment analysis method based on the influence of online stock reviews on stock market trends. This article takes online comments (posts) in Internet stock forums as the research object, including the amount of comment information and the emotional tendency expressed by specific content. In this paper, Chinese documents are processed for word segmentation. According to certain feature selection and feature extraction methods, feature words are selected to be counted as a feature word matrix, and the corresponding feature word frequency is calculated. The word segmentation tool used in this article is the Rwordseg package provided by R, through which the article can be easily segmented
to form a word frequency matrix. First, words are selected from the suggested word bank to match the second half of the stock review. If the matching is unsuccessful, the following words continue to be matched until the matching is successful and the matched words are used as the suggested words. This article divides the views of stock investors into rise, neutral and fall, and uses the trading day as the time unit to conduct sentiment analysis on the text data. While analyzing the weight of the stock review authors, this article also analyzes the attention degree of the stock reviews. The predicted value of the multi-data source stock prediction model of the simple sentiment index is closer to the actual closing price, the relative error between
the predicted value and the true value is no more than 1.9%, and the prediction trend accuracy rate is 78.9%, which does not include the simple sentiment index. The relative error between the predicted value and the true value of the multi-data source stock prediction model is 6.3%. The results show that online stock reviews affect the trend of stocks, positive sentiment causes stocks to rise slightly, and negative sentiment causes stocks to fall.

Keywords: Online Stock Reviews, Stock Market Trends, Text Sentiment Analysis, Stock Forecasts

Cite As

H. Li, W. Xue, "Impact of Online Stock Reviews on Stock Market Trends Using Text Sentiment Analysis Methods", 
Engineering Intelligent Systems, vol. 30 no. 6, pp. 431-440, 2022.




Downloads

Published

2022-11-01

Issue

Vol. 30 No. 6 (2022): International Journal of Engineering Intelligent Systems

Section

General Submission

License

The submission of a paper implies that, if accepted for publication, it will not be published elsewhere in the same form, in any language, without the prior consent of the publisher. Before publication, authors are requested to assign copyright to CRL Publishing Ltd. This allows CRL to sanction photocopying, and to authorize the reprinting of issues or volumes according to demand. Authors' traditional rights will not be jeopardized by assigning Copyright in this way, as they retain the right to reuse the material following publication, and to veto third-party publication.