用于快速不一致点挖掘的多维时间序列素描方法
Sketching Multidimensional Time Series for Fast Discord Mining
AI总结：
针对多维时间序列矩阵画像计算受维度制约的问题，提出一种维度无关的素描方法，在水处理和交通数据上实现最高50倍加速且质量损失极小，并支持维度动态增删。
摘要：
时间序列不一致点是时间序列异常检测中一种有用的基元，而矩阵画像能够有效捕捉不一致点。已有许多研究致力于提升不一致点发现相对于时间序列长度的可扩展性。然而，令人惊讶的是，很少有工作关注降低与多维时间序列维度相关的矩阵画像计算时间复杂度。在本工作中，我们提出了一种用于多维时间序列不一致点挖掘的素描方法。在与读取数据一样快的初始素描预处理之后，不一致点挖掘的运行时间与原始数据的维度无关。在来自水处理和交通运输的若干真实世界示例上，所提出的算法将吞吐量提升了至少一个数量级（50倍），并且对近似解的质量仅有极小影响。此外，所提出的方法能够以微不足道的开销处理维度的动态添加或删除。这使得数据分析师在探索数据时可以实时考虑“假设分析”场景。
英文摘要：
Time series discords are a useful primitive for time series anomaly detection, and the matrix profile is capable of capturing discord effectively. There exist many research efforts to improve the scalability of discord discovery with respect to the length of time series. However, there is surprisingly little work focused on reducing the time complexity of matrix profile computation associated with dimensionality of a multidimensional time series. In this work, we propose a sketch for discord mining among multi-dimensional time series. After an initial pre-processing of the sketch as fast as reading the data, the discord mining has runtime independent of the dimensionality of the original data. On several real world examples from water treatment and transportation, the proposed algorithm improves the throughput by at least an order of magnitude (50X) and only has minimal impact on the quality of the approximated solution. Additionally, the proposed method can handle the dynamic addition or deletion of dimensions inconsequential overhead. This allows a data analyst to consider "what-if" scenarios in real time while exploring the data.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
