<![CDATA[This study aims to design and implement a psychological testing website integrated with Large Language Models (LLMs) to map user psychological profiles and generate intelligent, personalized tourism recommendations in the Lagoi tourism area, Bintan, Indonesia. A Design and Development Research (DDR) approach was adopted, structured around the Waterfall model. The system was developed through requirement analysis, architectural design, and prototype implementation using ReactJS, Node.js/Express.js, MongoDB, and integrated with the Google Gemini-2.0-Flash API. Evaluation was conducted via expert validation and user testing. The system demonstrated high capability in analyzing non-linear psychological narratives using LLMs. It extracted travel-relevant personality traits with 90% relevance and 85% consistency (validated by psychologists). These profiles enabled personalized recommendations—such as Trikora Beach for peace-seekers and Treasure Bay for adventure enthusiasts. The average user satisfaction score was 4.15 (±0.72) on a 5-point scale, showing an 18.5% improvement over traditional demographic-based systems.Integrating LLMs with psychological profiling significantly enhances the personalization quality of tourism recommendation systems. This approach provides a scalable and adaptive solution for destination management to understand and serve diverse traveler segments more effectively.]]>
Copyrights © 2025
