Abstract:The similarity search on sensor data generated by a myriad of sensing devices is a frequently encountered problem in the era of the Internet of Things (IoT). This sensor data generally appear in the form of time series, a temporally ordered sequence of real numbers obtained regularly in time. It has been widely accepted that the dynamic time warping (DTW) currently is the most prevalent similarity measure in the time-series mining community, mainly due to its flexibility and broad applicability. However, calculating DTW between two time series has quadratic time complexity, leading to unsatisfactory efficiency when performing the similarity search over the large time-series data set. The main contribution of this article is to propose a method called product quantization (PQ)-based DTW (PQDTW) for fast time-series approximate similarity search under DTW. The PQ, a well-known approximate nearest neighbor search approach, is used in PQDTW. Nevertheless, the conventional PQ is developed with the Euclidean distance and is not designed for DTW. To solve this problem, the DTW barycenter averaging (DBA) technique is utilized to adapt the PQ for DTW before using it. We employ PQDTW along with the filter-and-refine framework to efficiently and accurately perform the time-series similarity search. Our method can reasonably reduce many DTW computations in the filtering phase; thus, the query process is accelerated. We compare PQDTW with related popular algorithms using public time-series data sets. Experimental results verify that the proposal achieves the best tradeoff between query efficiency and retrieval accuracy compared to the competitors.

Multilevel Dynamic Time Warping: A Parameter-Light Method for Fast Time Series Classification

Dynamic Time Warping under Product Quantization, with Applications to Time-Series Data Similarity Search

Piecewise Chebyshev Factorization Based Nearest Neighbour Classification for Time Series

Piecewise Factorization for Time Series Classification.

Integrating Data-Driven Segmentation, Local Feature Extraction and Fisher Kernel Encoding to Improve Time Series Classification

Dynamic Time Warping Constraint Learning for Large Margin Nearest Neighbor Classification.

Multivariate time series classification based on μσ-DWC feature and tree-structured M-SVM

Using dynamic time warping distances as features for improved time series classification

Learning a Mahalanobis Distance-Based Dynamic Time Warping Measure for Multivariate Time Series Classification

A Novel Distance Measure Based on Dynamic Time Warping to Improve Time Series Classification

Generalizing DTW to the multi-dimensional case requires an adaptive approach

Dynamic Time Warping: Itakura vs Sakoe-Chiba

Kernel Sparse Representation for Time Series Classification

TC-DTW: Accelerating Multivariate Dynamic Time Warping Through Triangle Inequality and Point Clustering

DTW-C++: Fast dynamic time warping and clustering of time series data

A Deep Multi-Task Representation Learning Method for Time Series Classification and Retrieval.

Exploiting Multi-Channels Deep Convolutional Neural Networks for Multivariate Time Series Classification

Generalized Time Warping Invariant Dictionary Learning for Time Series Classification and Clustering

Multitask Learning for Time Series Data with 2D Convolution

metricDTW: local distance metric learning in Dynamic Time Warping

Fast dynamic time warping and clustering in C++