2018/02/16 by Steven Van Vaerenbergh, Ignacio Santamarı́a, Van Vaerenbergh, Steven +5 · 1 citation
Computer Science · #Anomaly Detection Techniques and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Time Series Analysis and Forecasting
paper · pdf · doi:10.48550/arxiv.1802.05910
openalex publication_date 2018/02/16 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28
In this paper, we study the problem of locating a predefined sequence of\npatterns in a time series. In particular, the studied scenario assumes a\ntheoretical model is available that contains the expected locations of the\npatterns. This problem is found in several contexts, and it is commonly solved\nby first synthesizing a time series from the model, and then aligning it to the\ntrue time series through dynamic time warping. We propose a technique that\nincreases the similarity of both time series before aligning them, by mapping\nthem into a latent correlation space. The mapping is learned from the data\nthrough a machine-learning setup. Experiments on data from non-destructive\ntesting demonstrate that the proposed approach shows significant improvements\nover the state of the art.\n