Sentiment prediction model in social media data using beluga dodger optimization-based ensemble classifier

Twitter has developed into a significant social media network and is a topic of great interest for sentiment analysis researchers. Twitter is a crucial information source for learning about people's attitudes, feelings, viewpoints, and feedback. Every day, millions of reviews are written with both good and negative or indifferent feedback. The procedure of analyzing this generated review is difficult and time-consuming. This research utilized a beluga dodger optimization-based ensemble classifier to recognize and categorizes the sentiments in social media in order to overcome this problem. The ensemble classifier is combined with the beluga dodger (BD) optimization to create the classifier. Convolutional neural networks and bidirectional long short-term memory classifiers were combined to create the hybrid ensemble classifier, which performs more efficiently. In order to improve classification performance and obtain sentiment prediction more quickly, the proposed beluga dodger optimization started by mixing shark hunting characteristics with whale optimization. In comparison with all other approaches, the BD-optimized deep ensemble classifier achieved values of 97.61% accuracy, 96.20% sensitivity, and 98.61% specificity.