Business Analysis: User Attitude Evaluation and Prediction Based on Hotel User Reviews and Text Mining

In the post-pandemic era, the global hotel industry plays a crucial role in broader economic recovery, with consumer sentiment increasingly influencing market trends and operational strategies. This study utilizes advanced natural language processing (NLP) techniques and the Bidirectional Encoder Representations from Transformers (BERT) model to systematically analyze hotel user reviews, extracting profound insights into customer satisfaction and guiding targeted service improvements. By transforming unstructured textual reviews into high-dimensional feature vectors, the BERT model accurately classifies complex consumer emotions, uncovering nuanced patterns of satisfaction and dissatisfaction that traditional analytical methods often overlook. This sophisticated approach provides highly valuable, data-driven evidence for hotel management, helping them refine service offerings, optimize resource allocation, and significantly improve overall customer experiences. Furthermore, from a financial perspective, understanding consumer sentiment is vital for predicting market performance, as shifts in customer attitudes frequently correlate with stock price fluctuations and overall industry profitability. Additionally, the study rigorously addresses the pervasive issue of data imbalance in sentiment analysis by employing advanced techniques such as oversampling and undersampling to enhance model robustness and predictive accuracy. The empirical results offer actionable insights not only for hospitality practitioners but also for financial analysts, aiding in precise market forecasts and strategic investment decisions. Ultimately, this research highlights the transformative potential of deep learning-based sentiment analysis to drive sustainable business growth, improve financial outcomes, and enhance competitive advantage in the dynamic tourism and hospitality sectors, thereby contributing significantly to the broader economic landscape.