Moth Flame Optimization With Hybrid Deep Learning Based Sentiment Classification Toward ChatGPT on Twitter

ChatGPT, developed by OpenAI, is an advanced language model that excels at generating human-like text responses in conversational settings. As ChatGPT interacts with the user, it creates a range of sentiments from them, involving neutral, positive, or negative expressions. Sentiment analysis (SA), also called opinion mining, is a branch of natural language processing (NLP) that defines the emotional tone or sentiment conveyed in textual data. Sentiment analysis (SA) plays a major role in understanding how people respond and perceive different entities, involving services, products, brands, or artificial intelligence (AI) models GPT. Analyzing the sentiment toward ChatGPT gives valuable insight into, user experience, areas, and satisfaction for development. The study presents a moth flame optimization with hybrid deep learning-based sentiment analysis (MFOHDL-SA) on ChatGPT. The major aim of the MFOHDL-SA method is to design an automated AI model to properly classify the tweets as positive, negative, or neutral in sentiment towards ChatGPT. To accomplish this, the MFOHDL-SA technique initially pre-processes the tweets in different stages. Next, the TF-IDF model is used for the word embedding process. Moreover, the HDL method comprising a convolutional neural network with long short-term memory (CNN-LSTM) method was utilized for sentiment classification. To improve the classifier results of the HDL model, the MFO algorithm is used for hyperparameter tuning. The simulation results of the MFOHDL-SA technique are validated on the Twitter dataset from the Kaggle repository. The obtained experimental outcomes stated the betterment of the MFOHDL-SA approach over other existing techniques in terms of different measures. This provides a valued understanding of public sentiment towards ChatGPT on Twitter, allowing improved understanding and assessment of its impact and perception among users.