交互式多用户Skyline查询处理算法  

Algorithm for Interactive Multi-User Skyline Query

作　　者：邵路伊 秦小麟[1] 王潇逸 郭成盖 邓丹萍 SHAO Luyi, QIN Xiaolin, WANG Xiaoyi, GUO Chenggai, DENG Danping(College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, Chin)

机构地区：[1]南京航空航天大学计算机科学与技术学院,南京210016

出　　处：《计算机科学与探索》2018年第8期1202-1213,共12页Journal of Frontiers of Computer Science and Technology

基　　金：国家自然科学基金Nos.61373015;61402225;江苏省自然科学基金No.BK20140832~~

摘　　要：传统的Skyline查询是在单用户场景下设计的,随着数据库技术的发展以及应用新需求的出现,实际应用中考虑到多用户共同参与Skyline查询。在分析现有Skyline查询算法解决该问题不足的基础上,提出了一种基于权重的交互式多用户Skyline查询(weight-based interactive multi-user Skyline query,MUSW)算法。该算法定义了一种满意度度量方法,由用户权重决定Skyline结果集中每个数据点的满意度大小。MUSW算法旨在选取满意度大的数据点,首先确定多个子空间Skyline候选集;然后通过用户交互对用户权重进行动态调整,系统根据用户的反馈判断是否终止查询,从而使返回结果更符合用户的真实需求。在模拟和真实数据上验证MUSW算法的可行性,且具有良好的交互性能。The traditional Skyline query is designed in a single-user scenario. With the development of database technology and the emergence of new requirements, multiple users participate in Skyline query in the practical appli- cation. Based on the analysis of the existing Skyline query algorithms to solve the problem, this paper proposes a weight-based interactive multi-user Skyline query （MUSW） algorithm. The algorithm defines a satisfaction measure method that determines the satisfaction level of each data point in Skyline by the user＇s weight. The MUSW algo- rithm is designed to select the data points with high satisfaction. Firstly, the Skyline candidate sets are determined. Then, the user＇s weight is dynamically adjusted by user interaction with the system. The system judges whether to terminate the query according to the user＇s feedback, so that the return result is more in line with the user＇s real demand. The feasibility of MUSW algorithm is verified by simulation and real data, and it has good interactive performance.

关 键 词：多用户Skyline查询 交互查询 用户权重 动态调整 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]