Beyond Point Matching: Evaluating Multiscale Dubuc Distance for Time Series Similarity
Azim Ahmadzadeh, Mahsa Khazaei, Elaina Rohlfing
时间序列是高维度和复杂的数据对象,使得其高效的搜索和索引成为数据挖掘的长期挑战。 基于最近引入的相似性测量,即多尺度杜布距离(MDD),本文研究了其相对于广泛使用的动态时间翘曲(DTW)的相对优势和局限性。 MDD有两种关键方式新颖:它评估跨多个时间尺度的时间序列相似性,并避免点对点对齐。 我们证明,在许多MDD优于DTW的场景中,收益是可观的,我们对其解决的具体性能差距进行了详细分析。 除了来自UCR存档的95个数据集外,我们还提供模拟来测试我们的假设。 最后,我们将这两种方法应用于具有挑战性的现实世界分类任务,并表明MDD比DTW显着改进,强调了其实用性。
Time series are high-dimensional and complex data objects, making their efficient search and indexing a longstanding challenge in data mining. Building on a recently introduced similarity measure, namely Multiscale Dubuc Distance (MDD), this paper investigates its comparative strengths and limitations relative to the widely used Dynamic Time Warping (DTW). MDD is novel in two key ways: it evaluates time series similarity across multiple temporal scales and avoids point-to-point alignment. We dem...