CAJ | 학술논문

研究R树特点，考虑了离群点对R树结点构造的影响，结合改进的k-medoids 聚类算法提出了一种新的R树构造算法。与传统R树相比，新算法下构造的R树结点更加紧凑。通过实验证明，该优化算法构造的R树在查询性能方面的改进是明显的。
연구R수특점，고필료리군점대R수결점구조적영향，결합개진적k-medoids 취류산법제출료일충신적R수구조산법。여전통R수상비，신산법하구조적R수결점경가긴주。통과실험증명，해우화산법구조적R수재사순성능방면적개진시명현적。
This paper proposes a new method of constructing R-tree by studying its features, considering the outliers’ influence on constructing R-tree’s nodes and combining with improved k-medoids clustering algorithm. The R-tree based on this method has more compact structure compared with traditional R-tree. The results of the study show that, due to the optimization of structure, the improvement of R-tree in query efficiency is effectively.