Scholarlink

<p>The interconnectivity of global stock markets is flourishing, entailing the potential for vehemently intertwined stock price oscillations as a result of external disturbances, collaborative efforts, and competitive forces.Predicting the fluctuations of the stock market has been an age-old pursuit, where pioneers have employed conventional techniques to scrutinize the underlying principles and intricate technical elements.The market price of the shares varies indeterminately and is very hard to predict its final state at the end of the day. This creates havoc among the investors and creates a confusion state of taking decisions on the stocks of the companies. Occasionally, some erroneous prediction by the investors creates a huge loss to them by investing on wrong shares due to the wrong prediction. To overcome this concern and to predict the stock market status, this manuscript proposes a novel Ensemble Deep Learning Framework (EDLF), which employs deep learning algorithm for the high precision prediction of stock market data. The Deep Learning (DL) algorithm employs the hybrid Long Short Term Memory (LSTM) and Radial Basis Function Networks (RBFN) method for the prediction of stock market data. The proposed work exhibits a better performance with accuracy of 83.65%, 82.54% of precision, 83.94% of recall and a F1 score with 83.46%.</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

A Novel Stock Market Data Prediction Using Hybrid LSTM-RBFN Model

  • 期刊论文

  • 发表日期：

    2025年11月10日

查看全文
  • C. Selvan,
  • S. Sasikala,
  • S. Iwin Thanakumar Joseph,
  • V. Arulkumar,
  • S. Ranjana Devi

摘要

The interconnectivity of global stock markets is flourishing, entailing the potential for vehemently intertwined stock price oscillations as a result of external disturbances, collaborative efforts, and competitive forces.Predicting the fluctuations of the stock market has been an age-old pursuit, where pioneers have employed conventional techniques to scrutinize the underlying principles and intricate technical elements.The market price of the shares varies indeterminately and is very hard to predict its final state at the end of the day. This creates havoc among the investors and creates a confusion state of taking decisions on the stocks of the companies. Occasionally, some erroneous prediction by the investors creates a huge loss to them by investing on wrong shares due to the wrong prediction. To overcome this concern and to predict the stock market status, this manuscript proposes a novel Ensemble Deep Learning Framework (EDLF), which employs deep learning algorithm for the high precision prediction of stock market data. The Deep Learning (DL) algorithm employs the hybrid Long Short Term Memory (LSTM) and Radial Basis Function Networks (RBFN) method for the prediction of stock market data. The proposed work exhibits a better performance with accuracy of 83.65%, 82.54% of precision, 83.94% of recall and a F1 score with 83.46%.