사용자 선호도를 반영한 유사도 기반 Skyline Query 기법 = A Similarity-Based Skyline Query Scheme Reflecting User Preferences

다국어 초록 (Multilingual Abstract)

This study proposes a similarity-based skyline query method capable of quantitatively reflecting user preferences in multi-attribute recommendation environments. Traditional skyline approaches do not consider the relative importance of user preferences across attributes, and their computational cost increases exponentially as the number of dimensions grows, leading to performance degradation in filtering. To address these limitations, the proposed method applies Min- Max normalization to user-preferred attributes—such as distance, rating, and grade—and calculates the Euclidean distance between these attributes and the user preference vector to obtain a single similarity score. This similarity score is then combined with price, an independent attribute, to form a two-dimensional attribute vector (price, similarity). Based on this vector, the skyline query is executed to select alternative objects that align with user preferences. Experimental results demonstrate that the proposed method reduces the number of recommended objects by approximately 80–90% compared to the traditional approach, while also reducing the total number of operations to less than half, thereby simplifying the comparison structure and lowering computational cost. By integrating user-centric preference modeling with a dimension-reduction-based comparison framework, this study presents a filtering mechanism that is highly applicable to recommendation systems and real-time decision support environments. As a result, it can serve as a competitive solution in user-personalized recommendation scenarios where both precision and real-time processing efficiency are required.