Let's do the time warp again: non‐linear time series matching as a tool for sequentially structured data in ecology

• DOI: 10.1002/ecs2.3742
• Publication Date: 2021-09-10T01:37:55Z
• Authors: Jens C. Hegg; Brian P. Kennedy
• Keywords: Dynamic Time Warping; Data set; Temporal database

Abstract Ecological patterns are often fundamentally chronological. However, generalization of data is necessarily accompanied by a loss detail or resolution. Temporal in particular contain information not only values but the temporal structure, which lost when these aggregated to provide point estimates. Dynamic time warping (DTW) series comparison method that capable efficiently comparing despite offsets confound other methods. The DTW both efficient and remarkably flexible, matching also any sequentially structured set, has made it popular technique machine learning, artificial intelligence, big analytical tasks. rarely used ecology ubiquity temporally data. As technological advances have increased richness small‐scale ecological data, may be an attractive analysis because able utilize additional contained structure many sets. In this study, we use example set high‐resolution fish movement records obtained from otolith microchemistry compare traditional techniques with clustering. Our results suggest detecting subtle behavioral within sets aggregation cannot. These evidence useful across types commonly collected ecology, as well ordered “pseudo series” such classification species shape.