【摘要】 利用极大团把海量的数据项进行有效的划分,降低了后续数据挖掘和决策选择的复杂度.对于含有时间参量的原始数据,极大团具有一定的时域特性,挖掘其时间特性将进一步提高决策的准确度并可以减少分析数据的规模,降低对计算资源的需求.因此,在提出一种求极大有序频繁项目集算法的基础上,给出一种挖掘极大有序频繁项目集时间属性的方法.在时间并范围内实施搜索极大频繁项目集保证了搜索结果的无遗漏性,并以此为基础,通过定义频繁项目集关键时间段,较好地解释了极大频繁项目集的时间属性;通过实际数据验证了所给出方法的可行性和有效性.更多还原

【Abstract】 A maximum clique based mining algorithm reduces the complexity of information mining and decisions making.For the raw data with time features,discovering these characteristics shall further improve decisional accuracy and reduce request of computing resources.In this paper,we present a method to mine the maximum orderly frequent item sets from timed raw data,which assures that the constructed maximum orderly frequent item sets are complete.Moreover,key time interval of frequent item sets are introduced,and proved that it is a novel way to explain the time attribute of the frequent item sets better.The feasibility and effectiveness are verified through experiments.更多还原