Scholarlink

<p>In recent days, the evolving growth of social-media applications and their reviews have given rise to Sentimental analysis (SA) to analyze the attitudes, feelings, and views of the users. However, the traditional sentimental analysis mechanism possessed limitations in understanding the context of the text, generalization, interpretability, inaccurate analysis, and dialectal variation. Therefore, to address these aforementioned issues, the Ateles Leading Gorilla Optimizer enabled Deep Ensemble Activation Model (ALGO-DeAM) is proposed in this research. Specifically, the Ateles Leading Gorilla Optimization (ALGO) tunes the hyperparameters and selects the optimal features of the ALGO-DeAM model, which in turn accelerates the training process and minimizes the computation complexity. In real-time SA applications, this hybrid optimization approach offers enhanced accuracy, resilience, and efficiency, making it especially useful for processing high-dimensional and dynamic sentiment data. The DeAM takes advantage of various learning patterns and improves performance by capturing multiple aspects of the input. The proposed ALGO-DeAM attains higher performance with the metrics of accuracy, sensitivity, and specificity, as 98.09%, 96.91%, and 98.69% using the Sentiment140 dataset, 98.15%, 98.00%, and 98.43% with Twitter sentiment dataset, and 96.59%, 97.19%, and 97.15% with Emotion dataset respectively.</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

ALGO-DeAM: An improved Deep Ensemble Model for Twitter Sentimental Analysis using Ateles Leading Gorilla Optimizer

  • 期刊论文

  • 发表日期：

    2026年5月14日

查看全文
  • Supriya Sameer Nalawade,
  • Shamala R. Mahadik

摘要

In recent days, the evolving growth of social-media applications and their reviews have given rise to Sentimental analysis (SA) to analyze the attitudes, feelings, and views of the users. However, the traditional sentimental analysis mechanism possessed limitations in understanding the context of the text, generalization, interpretability, inaccurate analysis, and dialectal variation. Therefore, to address these aforementioned issues, the Ateles Leading Gorilla Optimizer enabled Deep Ensemble Activation Model (ALGO-DeAM) is proposed in this research. Specifically, the Ateles Leading Gorilla Optimization (ALGO) tunes the hyperparameters and selects the optimal features of the ALGO-DeAM model, which in turn accelerates the training process and minimizes the computation complexity. In real-time SA applications, this hybrid optimization approach offers enhanced accuracy, resilience, and efficiency, making it especially useful for processing high-dimensional and dynamic sentiment data. The DeAM takes advantage of various learning patterns and improves performance by capturing multiple aspects of the input. The proposed ALGO-DeAM attains higher performance with the metrics of accuracy, sensitivity, and specificity, as 98.09%, 96.91%, and 98.69% using the Sentiment140 dataset, 98.15%, 98.00%, and 98.43% with Twitter sentiment dataset, and 96.59%, 97.19%, and 97.15% with Emotion dataset respectively.