Scholarlink

<p>Stock market is a very delicate and volatile system which makes prediction extremely complex and difficult. However, with the rise of social media, conventional norms and systems are undergoing a massive change, Prediction of stock markets has always been a relevant field of study as traders and organisations are using deep learning (DL) techniques to improve their strategies. The impact of certain social media platforms such as Twitter on the stock market cannot be overstated. Thus, the given paper makes use of Bi-directional Long Short-Term Memory Recurrent Neural Network (BiLSTM-RNN) to predict short term fluctuations in the stock market with the help of tweets from the users. The tweets of stockholding accounts have been categorized as catalyst and irrelevant thereby flagging all the relevant tweets that can impact the market and warns the trader of upcoming changes. The model’s performance is assessed based on evaluation metrics such as root mean squared error (RMSE) and mean absolute percentage error (MAPE).</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Utilization of BiLSTM-RNN to identify tweets for forecasting stock market volatility

  • 期刊论文

  • 发表日期：

    2025年10月25日

查看全文
  • Ganesh Waghmare,
  • Jyoti Mishra,
  • Vijaya Amitesh Gondane,
  • Neetu Kumari,
  • Jairaj J. Sasane,
  • Ashok D. Todmal,
  • Padmakar Shahare,
  • Ravindra Khedkar

摘要

Stock market is a very delicate and volatile system which makes prediction extremely complex and difficult. However, with the rise of social media, conventional norms and systems are undergoing a massive change, Prediction of stock markets has always been a relevant field of study as traders and organisations are using deep learning (DL) techniques to improve their strategies. The impact of certain social media platforms such as Twitter on the stock market cannot be overstated. Thus, the given paper makes use of Bi-directional Long Short-Term Memory Recurrent Neural Network (BiLSTM-RNN) to predict short term fluctuations in the stock market with the help of tweets from the users. The tweets of stockholding accounts have been categorized as catalyst and irrelevant thereby flagging all the relevant tweets that can impact the market and warns the trader of upcoming changes. The model’s performance is assessed based on evaluation metrics such as root mean squared error (RMSE) and mean absolute percentage error (MAPE).