Predict Market Fluctuations Based on the TSI and the Sentiment of Financial Video News Sites via Machine Learning

2023 15th International Conference on Computer and Automation Engineering (ICCAE) Pub Date : 2023-03-03 DOI:10.1109/ICCAE56788.2023.10111493

Faten Alzazah, Xiaochun Cheng, Xiaohong Gao

{"title":"Predict Market Fluctuations Based on the TSI and the Sentiment of Financial Video News Sites via Machine Learning","authors":"Faten Alzazah, Xiaochun Cheng, Xiaohong Gao","doi":"10.1109/ICCAE56788.2023.10111493","DOIUrl":null,"url":null,"abstract":"Scientists have long been interested in forecasting stock market fluctuations. Traditional data like financial textual news, stock prices, and comments are simply no longer sufficient because they don't provide a comprehensive picture. In this study, the efficacy of using financial video news stories versus the use of conventional text news stories to forecast the stock market is examined. We used the Granger causality test to evaluate the robustness of the causal connection between share prices, text news sentiment, video news sentiments, and the Twitter sentiment index.Several models for sentiment analysis of S&P 500 stock were assessed using LR, SVM, LSTM, ATT-LSTM, and CNN models. This study is distinctive because it compares the use of financial video news stories, conventional text news stories, and the Twitter Sentiment Index to forecast stock market movements. The experimental results suggest that there is a stronger causal connection between video news sentiment and stock market fluctuation compared to conventional text news sentiments. The result shows that we can more accurately predict market changes using video news than we can with traditional news.","PeriodicalId":406112,"journal":{"name":"2023 15th International Conference on Computer and Automation Engineering (ICCAE)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 15th International Conference on Computer and Automation Engineering (ICCAE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCAE56788.2023.10111493","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Scientists have long been interested in forecasting stock market fluctuations. Traditional data like financial textual news, stock prices, and comments are simply no longer sufficient because they don't provide a comprehensive picture. In this study, the efficacy of using financial video news stories versus the use of conventional text news stories to forecast the stock market is examined. We used the Granger causality test to evaluate the robustness of the causal connection between share prices, text news sentiment, video news sentiments, and the Twitter sentiment index.Several models for sentiment analysis of S&P 500 stock were assessed using LR, SVM, LSTM, ATT-LSTM, and CNN models. This study is distinctive because it compares the use of financial video news stories, conventional text news stories, and the Twitter Sentiment Index to forecast stock market movements. The experimental results suggest that there is a stronger causal connection between video news sentiment and stock market fluctuation compared to conventional text news sentiments. The result shows that we can more accurately predict market changes using video news than we can with traditional news.

查看原文本刊更多论文

基于TSI和金融视频新闻网站情绪的机器学习预测市场波动

长期以来，科学家们一直对预测股市波动感兴趣。像金融文本新闻、股票价格和评论这样的传统数据已经不再足够了，因为它们不能提供全面的信息。在这项研究中，使用金融视频新闻故事与使用传统文本新闻故事来预测股票市场的功效进行了检验。我们使用格兰杰因果检验来评估股价、文本新闻情绪、视频新闻情绪和Twitter情绪指数之间因果关系的稳健性。使用LR、SVM、LSTM、at -LSTM和CNN模型对标准普尔500指数股票的几个情绪分析模型进行了评估。这项研究的独特之处在于，它比较了金融视频新闻故事、传统文本新闻故事和Twitter情绪指数的使用，以预测股市走势。实验结果表明，与传统的文字新闻情绪相比，视频新闻情绪与股市波动之间存在更强的因果关系。结果表明，利用视频新闻可以比传统新闻更准确地预测市场变化。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2023 15th International Conference on Computer and Automation Engineering (ICCAE)

自引率

0.00%

发文量