Scholarlink

<p>Share market price predictions are a hot topic of research in the financial field, which is a very crucial task that reduces investment risks and attracts investors and researchers of the enormous amount to invest for income. Even though existing researchers are focused on the aggregation of historical data to predict the stock price by various methods, it suffers from certain limitations such as computational complexity, data interpretability, scalability, and reliability. To address these challenges, an intelligent Herpestidae algorithm optimized with a modified Triplet attention-based Hybrid deep convolutional and Recurrent Neural Network (IHA-MTA-HD2RNN) is proposed effectively. The IHA-MTA-HD2RNN model enhances the research capability by reducing the gradient vanishing issues. Moreover, the proposed optimization provides enhanced robust validation with a simple structure and affords better data consistency to achieve efficient pre-defined values. Additionally, the complexity of the model is reduced by implementing the modified attention function, which evaluates the cross-dimensional interaction information about the data features and achieves accurate price prediction for sustainable development. Utilization of these components serve robust validation and offers better consistency to produce accurate results on Stock Market Price Prediction. The experimental research achieves a minimal error rate of Mean Absolute Error, Root Mean Square Error, Mean Square Error, and R2 of 10.71, 3.27, 7.69, 0.89 using BSE dataset and 9.71, 3.11, 6.66, and 0.89 using Nifty 50 dataset, when compared to state-of-the-art methods.</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

Stock market price prediction using intelligence herpestidae algorithm with modified triplet attention-based hybrid deep learning model

  • 期刊论文

  • 发表日期：

    2026年4月17日

查看全文
  • Pranjali Kasture,
  • Kamini Shirsath

摘要

Share market price predictions are a hot topic of research in the financial field, which is a very crucial task that reduces investment risks and attracts investors and researchers of the enormous amount to invest for income. Even though existing researchers are focused on the aggregation of historical data to predict the stock price by various methods, it suffers from certain limitations such as computational complexity, data interpretability, scalability, and reliability. To address these challenges, an intelligent Herpestidae algorithm optimized with a modified Triplet attention-based Hybrid deep convolutional and Recurrent Neural Network (IHA-MTA-HD2RNN) is proposed effectively. The IHA-MTA-HD2RNN model enhances the research capability by reducing the gradient vanishing issues. Moreover, the proposed optimization provides enhanced robust validation with a simple structure and affords better data consistency to achieve efficient pre-defined values. Additionally, the complexity of the model is reduced by implementing the modified attention function, which evaluates the cross-dimensional interaction information about the data features and achieves accurate price prediction for sustainable development. Utilization of these components serve robust validation and offers better consistency to produce accurate results on Stock Market Price Prediction. The experimental research achieves a minimal error rate of Mean Absolute Error, Root Mean Square Error, Mean Square Error, and R2 of 10.71, 3.27, 7.69, 0.89 using BSE dataset and 9.71, 3.11, 6.66, and 0.89 using Nifty 50 dataset, when compared to state-of-the-art methods.