Abstract

Reviews are the most crucial opinions delivered by individuals based on their experiences with a product or service. Analyzing reviews aims to boost brand reputation, ensure quality, earn customer trust, pinpoint areas for improvement, and differentiate a business from its competitors. The study aims to explore individual perspectives on emotional tendencies by employing a deep learning-based text mining approach called opinion mining, on the Disneyland reviews dataset. General Data cleansing techniques are applied to eliminate irrelevant or noisy information. To analyze perception this study utilized the Text Blob package to extract polarities of the text, phrases, or words. Furthermore, this study employed a sophisticated K-Fold Cross Validated LSTM model, which leverages deep learning to classify sentiments and provide predictive analysis. In short, the model is a type of artificial intelligence that can understand and interpret emotions or opinions expressed in text data through the LSTM network and the cross-validation technique. By integrating the strengths of both methods, the proposed model enables the identification of people’s perceptions and ensures performance stability. The findings show that model performance resulted in an average accuracy of 0.9435 and f1-score of 0.936. In a comprehensive evaluation of abilities, the K-fold LSTM technique outperformed traditional machine learning methods.