<![CDATA[The rapid growth of digital technologies has transformed the tourism industry and increased the need for personalized recommendation systems to enhance user experience and business competitiveness. However, many small- and medium-scale travel agencies still rely on manual reservation processes and social media–based promotions, which limit service efficiency and personalization. This study designs and implements a web-based reservation and tourism recommendation system for Sumovacation Tour using a Content-Based Filtering approach enhanced with feature weighting and cosine similarity. The main novelty of this study lies in the feature weighting mechanism, which assigns different importance levels to package attributes such as activities, travel duration, package type, and budget, improving recommendation relevance compared to standard content-based methods. Data were collected from Google Maps reviews in 2025, resulting in approximately 300 rating and review entries. The recommendation engine computes weighted relevance scores from user preference tags and package metadata to generate personalized recommendations. System functionality was validated using Black Box Testing, where all core workflows successfully passed, while usability evaluation using the USE Questionnaire showed high user acceptance, with usefulness, satisfaction, and ease of use each scoring 94.4%, and ease of learning reaching 95.2%. During testing, challenges related to data consistency and user input variation were addressed through input validation. The results show that the proposed system improves recommendation relevance while enhancing operational efficiency by reducing manual booking handling and supporting digital reservation management.]]>
