Scholarlink

<p>Sentiment analysis is one of the most widely used methods for forecasting stock market action from consumer feedback. Most of the methods associated with sentiment analysis are limited due to low accuracy and enhanced error rate. This is addressed by proposing a synergic squeeze deep info convolutional neural network-advanced variable capsule equilibrium auto encoder (SSDCNN-AVCEAE) for sentiment analysis and accurate multiple data-based stock prediction. Stock market data from NSE Nifty 50 (Mar 2, 2020–May 10, 2021) and real-time twitter sentiment analysis are pre-processed through data cleaning and sentiment analyzer lexicon processes. Merging features using SSDCNN, optimized with random search algorithm, mitigates overfitting. SSDCNN eliminates redundant features. Selected features undergo classification by AVCEAE, a fusion of advanced capsule auto encoder (ACAE) and variable equilibrium optimization algorithm, enhancing prediction accuracy for rising or falling stock market movements while minimizing errors. Variable equilibrium optimization refines ACAE parameters. The proposed framework demonstrates exceptional performance with F1-Score, accuracy, false alarm rate, sensitivity, precision, specificity, and error rate reaching 98%, 99%, 0.1%, 99%, 99%, and 0.2%, respectively. The measurements highlight the model's ability to handle a variety of issues, making it a reliable option for precise stock prediction.</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

An Accurate Multiple Data Based Stock Prediction and Sentiment Analysis Using Synergic Deep Info Convolutional Neural Network

  • 期刊论文

  • 发表日期：

    2025年4月21日

查看全文
  • T. M. Sanara,
  • M. Umme Salma

摘要

Sentiment analysis is one of the most widely used methods for forecasting stock market action from consumer feedback. Most of the methods associated with sentiment analysis are limited due to low accuracy and enhanced error rate. This is addressed by proposing a synergic squeeze deep info convolutional neural network-advanced variable capsule equilibrium auto encoder (SSDCNN-AVCEAE) for sentiment analysis and accurate multiple data-based stock prediction. Stock market data from NSE Nifty 50 (Mar 2, 2020–May 10, 2021) and real-time twitter sentiment analysis are pre-processed through data cleaning and sentiment analyzer lexicon processes. Merging features using SSDCNN, optimized with random search algorithm, mitigates overfitting. SSDCNN eliminates redundant features. Selected features undergo classification by AVCEAE, a fusion of advanced capsule auto encoder (ACAE) and variable equilibrium optimization algorithm, enhancing prediction accuracy for rising or falling stock market movements while minimizing errors. Variable equilibrium optimization refines ACAE parameters. The proposed framework demonstrates exceptional performance with F1-Score, accuracy, false alarm rate, sensitivity, precision, specificity, and error rate reaching 98%, 99%, 0.1%, 99%, 99%, and 0.2%, respectively. The measurements highlight the model's ability to handle a variety of issues, making it a reliable option for precise stock prediction.