Natural language processing and intelligent decision optimization technology for user feedback in hotel management

With the deep penetration of artificial intelligence technology in the hotel industry, natural language processing and reinforcement learning algorithms offer new approaches to solving the issues of inefficient user feedback analysis and lagging decision optimization in hotel management. Addressing the characteristics of hotel user feedback, specifically its colloquial nature and semantic complexity, and the inability of traditional manual analysis methods to meet the demands of large-scale, high-timeliness analysis, this study proposes a novel approach. It employs a Bidirectional Encoder Representations from Transformer model based on whole-word masking to enhance the semantic representation capabilities of colloquial Chinese feedback. This is combined with a Bidirectional Long Short-Term Memory network to capture long-range temporal dependencies within the text context. Finally, by integrating Deep Deterministic Policy Gradient, an intelligent optimization decision-making module is constructed. The synthesis of these three components establishes an intelligent optimization technology for hotel management driven by user feedback. The comparative experiment results showed that the technology outperformed the comparative model in terms of hotel user feedback analysis and core indicators of management decision optimization. The accuracy of sentiment classification was 92.37%, the accuracy of problem extraction was 89.64%, and the final accuracy of positive, neutral, and negative sentiment text classification was stable at 93.5%, 89%, and 92.5%, respectively. From this, the proposed technology can capture the core semantics and problem characteristics of hotel user feedback, enhance the scientific nature of hotel management decisions and the precision of service optimization, and provide a scientific basis for the intelligent and data-driven transformation of the hotel industry, as well as word-of-mouth improvement and customer retention.