Aquila Optimization Algorithm with Advanced Learning Model-Based Sentiment Analysis on Social Media Environment

Abstract

Sentiment Analysis (SA) on social media is a text mining method that includes employing Natural Language Processing (NLP), and Machine Learning (ML) approaches to categorize and assess the opinions, attitudes, and emotional tone expressed in user-generated content on platforms including Instagram, Twitter, and Facebook. This analysis provides meaningful information for researchers, marketers, and businesses to track trends, gauge public sentiment, and make data-driven decisions by automatically classifying text as positive, negative, or neutral, which helps them understand brand perception, customer satisfaction, and emerging problems in the dynamic world of online communication. Deep Learning (DL) based SA on social networking media is a robust NLP tool that leverages neural networks, including Recurrent Neural Network (RNN) or transformer models such as GPT and BERT, to automatically define the sentiment expressed in user-generated content on platforms like Facebook, Twitter, and Instagram. This manuscript offers an Aquila Optimization Algorithm with Deep Learning based Sentiment Analysis and Classification (AOADL-SAC) technique on social networking. The AOADL-SAC technique aims to recognize and classify the sentiments on social media. In the presented AOADL-SAC technique, pre-processing was implemented to convert the data provided as input into a compatible format. In addition, the AOADL-SAC technique implements a Long Short-Term Memory (LSTM) approach for recognizing and classifying sentiments. At last, the AOA-based tuning procedure was performed to adjust the hyperparameters of the LSTM approach. The investigational output of the AOADL-SAC technique is examined on a standard dataset. The comprehensive outputs highlighted the AOADL-SAC method and stated the AOADL-SAC technique accomplishes better outcomes than recent techniques concerning distinct aspects.