Mining Structural Changes

矿业结构变化

• 批准号: 21300047
• 负责人: HARAGUCHI Makoto
• 金额: $ 4.16万
• 依托单位: Hokkaido University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 2009
• 资助国家: 日本
• 起止时间: 2009 至 2011
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-21300047/
• 关键词: 変化検出問題; 情報量差分; 制約マイニング; 情報量変化; クリーク枚挙; 組み合わせ最適化; クラスタリング; クラスター構造; 分枝限定法; クリーク制約; Emerging Pattern; パターンの摂動; 稀少パターン発見

We have targeted patterns of variables whose correlations get increased after an event, while the correlations are uncorrelated before the event. To take into account positive, negative and even partial correlations among variables, we adopt Kullback-Leibler divergence for two contexts, before and after the event, and take their difference as the measure of changes. To reduce the computational cost for set partitioning, we implemented two strategies, Double Clique Constraints and Jumping Emerging Correlation Change, and showed that the correlation change problem can be solved efficiently even for large data sets.

我们已经针对变量的目标模式，其相关性在事件发生后增加，而事件发生前的相关性是不相关的。为了考虑变量之间的正相关，负相关甚至部分相关性，我们在事件发生前和之后采用Kullback-Leibler差异，并将它们的区别作为变化的度量。为了降低设定分区的计算成本，我们实施了两种策略，双重集团的约束和跳跃新兴的相关性变化，并表明相关性变化问题即使对于大型数据集也可以有效地解决。

项目成果

Top-N Minimization Approach for Indicative Correlation Change Mining

• DOI: 10.1007/978-3-642-31537-4_9
• 发表时间: 2012-07
• 作者: Aixiang Li;M. Haraguchi;Yoshiaki Okubo

Finding Top-N Pseudo Formal Concepts with Core Intents

寻找具有核心意图的前 N ​​个伪形式概念

• 发表时间: 2009
• 期刊: Proc.of the 6th International Conference on Machine Learning and Data Mining-MLDM'09 Springer-LNAI 5632
• 作者: Y.Okubo;M.Haraguchi
• 通讯作者: M.Haraguchi

Finding Rare Patterns with Weak Correlation Constraint

寻找具有弱相关约束的罕见模式

• 发表时间: 2010
• 期刊: Proc. of the2010 IEEE Int'l Conf. on Data MiningWorkshops-ICDMW'10
• 作者: Y. Okubo;M. Haraguchi and T. Nakajima
• 通讯作者: M. Haraguchi and T. Nakajima

Pinpoint Clustering of Web Pages and Mining Implicit Crossover Concepts

• DOI: 10.5772/8380
• 发表时间: 2010-03
• 作者: M. Haraguchi;Yoshiaki Okubo

Contrasting Correlations by an Efficient Double-Clique Condition

通过有效的双团条件对比相关性

• 发表时间: 2012
• 期刊: Transactions on Machine Learning and Data Mining
• 作者: Li Aixiang;Makoto Haraguchi and Yoshiaki Okubo
• 通讯作者: Makoto Haraguchi and Yoshiaki Okubo