Previous weighted frequent pattern (WFP) mining algorithms are not suitable for data streams for they need multiple database scans. In this paper, we present an efficient algorithm SWFP-Miner to mine weighted freque...Previous weighted frequent pattern (WFP) mining algorithms are not suitable for data streams for they need multiple database scans. In this paper, we present an efficient algorithm SWFP-Miner to mine weighted frequent pattern over data streams. SWFP-Miner is based on sliding window and can discover important frequent pattern from the recent data. A new refined weight definition is proposed to keep the downward closure property, and two pruning strategies are presented to prune the weighted infrequent pattern. Experimental studies are performed to evaluate the effectiveness and efficiency of SWFP-Miner.展开更多
针对无线传感器网络探测网络环境的自适应休眠算法(Probing Environment and Adaptive Sleeping,PEAS)在节点调度过程中,存在节点能耗不均衡、网络的生命周期较短的问题,提出一种基于加权的优化覆盖算法。该算法对最小频繁项的目标所对...针对无线传感器网络探测网络环境的自适应休眠算法(Probing Environment and Adaptive Sleeping,PEAS)在节点调度过程中,存在节点能耗不均衡、网络的生命周期较短的问题,提出一种基于加权的优化覆盖算法。该算法对最小频繁项的目标所对应的传感节点按能量高低进行划分集合,使各集合能够独立覆盖最小频繁项的目标,以达到局部的优化。考虑到传感节点覆盖目标数和剩余能量对无线传感网络生存周期的影响,对边缘未覆盖的目标节点采用加权的方式进行覆盖。仿真结果表明:该算法能够均衡网络节点的能耗,有效地延长了网络的生命周期。展开更多
提出一种矩阵加权关联模式支持度计算方法及其相关定理,给出矩阵加权项集剪枝策略,基于该剪枝策略提出一种基于项权值变化的矩阵加权关联规则挖掘算法MWAR-Miner(matrix-weighted association rules-miner)。该算法克服现有的项无加权...提出一种矩阵加权关联模式支持度计算方法及其相关定理,给出矩阵加权项集剪枝策略,基于该剪枝策略提出一种基于项权值变化的矩阵加权关联规则挖掘算法MWAR-Miner(matrix-weighted association rules-miner)。该算法克服现有的项无加权和项权值固定条件下挖掘关联规则的缺陷,采用新的剪枝技术和模式支持度计算方法挖掘有效的矩阵加权关联规则,避免无效的和无趣的模式产生。以中文数据集CWT200g和英文数据集NTCIR-5为实验数据,理论分析和实验结果表明,与现有矩阵加权模式挖掘算法和基于无加权的挖掘算法比较,该算法挖掘的候选项集数量和挖掘时间明显减少,挖掘效率得到极大提高。展开更多
文摘Previous weighted frequent pattern (WFP) mining algorithms are not suitable for data streams for they need multiple database scans. In this paper, we present an efficient algorithm SWFP-Miner to mine weighted frequent pattern over data streams. SWFP-Miner is based on sliding window and can discover important frequent pattern from the recent data. A new refined weight definition is proposed to keep the downward closure property, and two pruning strategies are presented to prune the weighted infrequent pattern. Experimental studies are performed to evaluate the effectiveness and efficiency of SWFP-Miner.
文摘针对无线传感器网络探测网络环境的自适应休眠算法(Probing Environment and Adaptive Sleeping,PEAS)在节点调度过程中,存在节点能耗不均衡、网络的生命周期较短的问题,提出一种基于加权的优化覆盖算法。该算法对最小频繁项的目标所对应的传感节点按能量高低进行划分集合,使各集合能够独立覆盖最小频繁项的目标,以达到局部的优化。考虑到传感节点覆盖目标数和剩余能量对无线传感网络生存周期的影响,对边缘未覆盖的目标节点采用加权的方式进行覆盖。仿真结果表明:该算法能够均衡网络节点的能耗,有效地延长了网络的生命周期。
文摘提出一种矩阵加权关联模式支持度计算方法及其相关定理,给出矩阵加权项集剪枝策略,基于该剪枝策略提出一种基于项权值变化的矩阵加权关联规则挖掘算法MWAR-Miner(matrix-weighted association rules-miner)。该算法克服现有的项无加权和项权值固定条件下挖掘关联规则的缺陷,采用新的剪枝技术和模式支持度计算方法挖掘有效的矩阵加权关联规则,避免无效的和无趣的模式产生。以中文数据集CWT200g和英文数据集NTCIR-5为实验数据,理论分析和实验结果表明,与现有矩阵加权模式挖掘算法和基于无加权的挖掘算法比较,该算法挖掘的候选项集数量和挖掘时间明显减少,挖掘效率得到极大提高。