摘要
不确定数据上的概率相互最近邻查询具有重要的实际应用,针对目前关于这方面的研究尚少,提出了不确定数据上的概率相互最近邻的top-k查询算法。首先对问题进行描述与定义,其次总结可行的裁剪规则,从而裁剪查询对象中未计算的实例点。通过实验表明,该算法能有效地降低最近邻查询中的I/O开销,提高查询的响应速度。
Probabilistic mutual nearest neighbors query has important practical applications,but there are few researches about this problem so far.This paper proposed an algorithm to find the top-k probabilistic mutual nearest neighbors on uncertain data.First,described and formalized the problem.Second,proposed some pruning rules to cut some instances that were not need to be accessed in the query object.As the experimental results show that this algorithm can reduce the I/O cost effectively,and improve the response speed.
出处
《计算机应用研究》
CSCD
北大核心
2011年第7期2485-2487,2491,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(60721002)
国家"863"计划资助项目(2007AA01Z178)
关键词
空间查询
不确定数据
概率相互最近邻
反向最近邻
查询处理
spatial query
uncertain data
mutual nearest neighbor
reverse nearest neighbor
query processing