<![CDATA[This study investigates the enhancement of tourism destination recommendation systems through the use of K-Means clustering to improve training data quality and model accuracy. The rapid advancement of information technology has increased the demand for personalized and accurate recommendation systems within the tourism industry. Despite this, achieving high prediction accuracy remains a significant challenge. This study employs K-Means clustering to segment training data into homogeneous clusters, thereby improving data representation and enhancing the predictive accuracy of recommendation models. The research methodology includes a comprehensive literature review, data collection, preprocessing, clustering, and model testing using K-Nearest Neighbors (KNN), Decision Tree, and Naive Bayes algorithms. The results show that after applying K-Means clustering, KNN's accuracy increased by 2.27%, and its kappa and precision values also improved, indicating enhanced reliability and prediction accuracy. Naive Bayes exhibited substantial improvements with a 9.09% increase in accuracy, alongside significant enhancements in kappa and precision metrics. Conversely, the Decision Tree algorithm experienced a decline in performance after clustering. Therefore, clustering techniques are not suitable for application to the Decision Tree algorithm.]]>
Copyrights © 2026
