一种基于上界值剪枝的用户敏感top-k dominating查询方法  

User-sensitive top-k dominating query method based on upper bound pruning

作　　者：徐超 吴丹琪[1,2] 陈勇 雷锦涛 Xu Chao;Wu Danqi;Chen Yong;Lei Jintao(School of Information Engineering,Nanjing Audit University,Nanjing 211815,China;Institute of Audit Science&Technology,Nanjing Audit University,Nanjing 211815,China)

机构地区：[1]南京审计大学信息工程学院,南京211815 [2]南京审计大学审计科学与技术研究院,南京211815

出　　处：《计算机应用研究》2023年第1期198-203,208,共7页Application Research of Computers

基　　金：国家自然科学基金资助项目(71972102);教育部人文社会科学研究规划基金资助项目(19YJAZH100);江苏省高等学校自然科学研究重大项目(20KJA520002);江苏省高校优秀科技创新团队(2021)。

摘　　要：信息化社会带来了数据量的快速增长,但也导致数据的完整性和价值密度越来越低,如何从海量不完整数据中快速找到有价值的信息用于用户个性化推荐,已经成为研究的热点。结合上界值剪枝算法,提出一种用户敏感top-k dominating(TKD)查询方法(USTD)。该方法首先结合用户兴趣度,重新定义了不完整数据集上对象的支配关系及其度量方法,计算对象的权重支配分数;然后以重新定义的支配关系,证明了依据上界值可剪枝的条件;最后将上界值剪枝条件融于USTD的查询过程,以从数据集中快速找到前k个满足用户兴趣的数据。实验结果表明,USTD算法在查询速度上有一定提升,在查询结果的评分上具有显著优势。With the rapid growth of data volume in the information society, the integrity and value density of data are getting lower and lower, but people’s personalized needs are becoming more and more obvious.How to quickly find valuable data from massive incomplete data for users’ personalized recommendation has became a hot spot of research.Combined with the upper bound pruning algorithm, this paper proposed a user-sensitive top-k dominating(TKD) query method(USTD).Firstly, combined with user’s interests, the method redefined the dominance relationship and its measurement method of objects on incomplete datasets, and calculate the weight dominance grade of the object.Then, it proved the condition of pruning according to the upper bound value by using the redefined dominance relationship.Finally, it integrated the upper bound pruning condition into the query process of USTD to quickly find the first k data satisfying users’ interests from the dataset.The experimental results show that the USTD algorithm has a certain improvement in query speed and a significant advantage in the scoring of query results.

关 键 词：不完整数据 TKD查询 用户敏感 权重支配分数 上界值剪枝 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]