Exact indexing of dynamic time warping

The problem of indexing time series has attracted much interest. Most algorithms used to index time series utilize the Euclidean distance or some variation thereof. However, it has been forcefully shown that the Euclidean distance is a very brittle distance measure. Dynamic time warping (DTW) is a much more robust distance measure for time series, allowing similar shapes to match even if they are out of phase in the time axis. Because of this flexibility, DTW is widely used in science, medicine, industry and finance. Unfortunately, however, DTW does not obey the triangular inequality and thus has resisted attempts at exact indexing. Instead, many researchers have introduced approximate indexing techniques or abandoned the idea of indexing and concentrated on speeding up sequential searches. In this work, we introduce a novel technique for the exact indexing of DTW. We prove that our method guarantees no false dismissals and we demonstrate its vast superiority over all competing approaches in the largest and most comprehensive set of time series indexing experiments ever undertaken.     

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

Dynamic time warping (DTW) has proven itself to be an exceptionally strong distance measure for time series. DTW in combination with one-nearest neighbor, one of the simplest machine learning methods, has been difficult to convincingly outperform on the time series classification task. In this paper, we present a simple technique for time series classification that exploits DTW’s strength on this task. But instead of directly using DTW as a distance measure to find nearest neighbors, the technique uses DTW to create new features which are then given to a standard machine learning method. We experimentally show that our technique improves over one-nearest neighbor DTW on 31 out of 47 UCR time series benchmark datasets. In addition, this method can be easily extended to be used in combination with other methods. In particular, we show that when combined with the symbolic aggregate approximation (SAX) method, it improves over it on 37 out of 47 UCR datasets. Thus the proposed method also provides a mechanism to combine distance-based methods like DTW with feature-based methods like SAX. We also show that combining the proposed classifiers through ensembles further improves the performance on time series classification.     

Weighted dynamic time warping for time series classification

Dynamic time warping (DTW), which finds the minimum path by providing non-linear alignments between two time series, has been widely used as a distance measure for time series classification and clustering. However, DTW does not account for the relative importance regarding the phase difference between a reference point and a testing point. This may lead to misclassification especially in applications where the shape similarity between two sequences is a major consideration for an accurate recognition. Therefore, we propose a novel distance measure, called a weighted DTW (WDTW), which is a penalty-based DTW. Our approach penalizes points with higher phase difference between a reference point and a testing point in order to prevent minimum distance distortion caused by outliers. The rationale underlying the proposed distance measure is demonstrated with some illustrative examples. A new weight function, called the modified logistic weight function (MLWF), is also proposed to systematically assign weights as a function of the phase difference between a reference point and a testing point. By applying different weights to adjacent points, the proposed algorithm can enhance the detection of similarity between two time series. We show that some popular distance measures such as DTW and Euclidean distance are special cases of our proposed WDTW measure. We extend the proposed idea to other variants of DTW such as derivative dynamic time warping (DDTW) and propose the weighted version of DDTW. We have compared the performances of our proposed procedures with other popular approaches using public data sets available through the UCR Time Series Data Mining Archive for both time series classification and clustering problems. The experimental results indicate that the proposed approaches can achieve improved accuracy for time series classification and clustering problems.     

Extensions and relationships of some existing lower-bound functions for dynamic time warping

Dynamic time warping (DTW) is a state-of-the-art time series similarity measure method, which warps time axes to match the same shape between two time series with different lengths. However, its quadratic time and space complexity is an obstacle to its applications in the large time series data mining. To address this problem, some lower-bound functions for DTW, fast methods to approximately measure the distance between time series, are used to prune the dissimilar objects from time series database so as to retain the candidates for further measuring their similarity with DTW. So far, the existing lower-bound functions for DTW have been widely accepted for time series similarity search and indexing. In this paper, we propose the extensions of two existing lower-bound functions and discuss the relationships among them. The extensions are improved with high tightness and without much time cost. At the same time, we theoretically prove that these extensions satisfy lower-bound requirement and are better than their old versions respectively. The experimental results demonstrate that in most cases the quality of the proposed extensions of lower-bound functions for DTW outperforms the original versions except for a slightly higher time cost.     

Fast and Exact Warping of Time Series Using Adaptive Segmental Approximations

Similarity search is a core module of many data analysis tasks, including search by example, classification, and clustering. For time series data, Dynamic Time Warping (DTW) has been proven a very effective similarity measure, since it minimizes the effects of shifting and distortion in time. However, the quadratic cost of DTW computation to the length of the matched sequences makes its direct application on databases of long time series very expensive. We propose a technique that decomposes the sequences into a number of segments and uses cheap approximations thereof to compute fast lower bounds for their warping distances. We present several, progressively tighter bounds, relying on the existence or not of warping constraints. Finally, we develop an index and a multi-step technique that uses the proposed bounds and performs two levels of filtering to efficiently process similarity queries. A thorough experimental study suggests that our method consistently outperforms state-of-the-art methods for DTW similarity search.     

Speeding up dynamic time warping distance for sparse time series data

Dynamic time warping (DTW) distance has been effectively used in mining time series data in a multitude of domains. However, in its original formulation DTW is extremely inefficient in comparing long sparse time series, containing mostly zeros and some unevenly spaced nonzero observations. Original DTW distance does not take advantage of this sparsity, leading to redundant calculations and a prohibitively large computational cost for long time series. We derive a new time warping similarity measure (AWarp) for sparse time series that works on the run-length encoded representation of sparse time series. The complexity of AWarp is quadratic on the number of observations as opposed to the range of time of the time series. Therefore, AWarp can be several orders of magnitude faster than DTW on sparse time series. AWarp is exact for binary-valued time series and a close approximation of the original DTW distance for any-valued series. We discuss useful variants of AWarp: bounded (both upper and lower), constrained, and multidimensional. We show applications of AWarp to three data mining tasks including clustering, classification, and outlier detection, which are otherwise not feasible using classic DTW, while producing equivalent results. Potential areas of application include bot detection, human activity classification, search trend analysis, seismic analysis, and unusual review pattern mining.     

Boundary-based lower-bound functions for dynamic time warping and their indexing

Dynamic time warping (DTW) is a powerful technique in the time-series similarity search. However, its performance on large-scale data is unsatisfactory because of its high computational cost and the fact that it cannot be indexed directly. The lower bound technique for DTW is an effective solution to this problem. In this paper, we explain the existing lower-bound functions from a unified perspective and show that they are only special cases under our framework. We then propose a group of lower-bound functions for DTW and compare their performances through extensive experiments. The experimental results show that the new methods are better than the existing ones in most cases, and a theoretical explanation of the results is also given. We further implement an index structure based on the new lower-bound function. Experimental results demonstrate a similar conclusion.     

Haar wavelets for efficient similarity search of time-series: with and without time warping

We address the handling of time series search based on two important distance definitions: Euclidean distance and time warping distance. The conventional method reduces the dimensionality by means of a discrete Fourier transform. We apply the Haar wavelet transform technique and propose the use of a proper normalization so that the method can guarantee no false dismissal for Euclidean distance. We found that this method has competitive performance from our experiments. Euclidean distance measurement cannot handle the time shifts of patterns. It fails to match the same rise and fall patterns of sequences with different scales. A distance measure that handles this problem is the time warping distance. However, the complexity of computing the time warping distance function is high. Also, as time warping distance is not a metric, most indexing techniques would not guarantee any false dismissal. We propose efficient strategies to mitigate the problems of time warping. We suggest a Haar wavelet-based approximation function for time warping distance, called Low Resolution Time Warping, which results in less computation by trading off a small amount of accuracy. We apply our approximation function to similarity search in time series databases, and show by experiment that it is highly effective in suppressing the number of false alarms in similarity search.     

Correlation based dynamic time warping of multivariate time series

In recent years, dynamic time warping (DTW) has begun to become the most widely used technique for comparison of time series data where extensive a priori knowledge is not available. However, it is often expected a multivariate comparison method to consider the correlation between the variables as this correlation carries the real information in many cases. Thus, principal component analysis (PCA) based similarity measures, such as PCA similarity factor (SPCA), are used in many industrial applications.In this paper, we present a novel algorithm called correlation based dynamic time warping (CBDTW) which combines DTW and PCA based similarity measures. To preserve correlation, multivariate time series are segmented and the local dissimilarity function of DTW originated from SPCA. The segments are obtained by bottom-up segmentation using special, PCA related costs. Our novel technique qualified on two databases, the database of signature verification competition 2004 and the commonly used AUSLAN dataset. We show that CBDTW outperforms the standard SPCA and the most commonly used, Euclidean distance based multivariate DTW in case of datasets with complex correlation structure.     

iSAX: disk-aware mining and indexing of massive time series datasets

Current research in indexing and mining time series data has produced many interesting algorithms and representations. However, the algorithms and the size of data considered have generally not been representative of the increasingly massive datasets encountered in science, engineering, and business domains. In this work, we introduce a novel multi-resolution symbolic representation which can be used to index datasets which are several orders of magnitude larger than anything else considered in the literature. To demonstrate the utility of this representation, we constructed a simple tree-based index structure which facilitates fast exact search and orders of magnitude faster, approximate search. For example, with a database of one-hundred million time series, the approximate search can retrieve high quality nearest neighbors in slightly over a second, whereas a sequential scan would take tens of minutes. Our experimental evaluation demonstrates that our representation allows index performance to scale well with increasing dataset sizes. Additionally, we provide analysis concerning parameter sensitivity, approximate search effectiveness, and lower bound comparisons between time series representations in a bit constrained environment. We further show how to exploit the combination of both exact and approximate search as sub-routines in data mining algorithms, allowing for the exact mining of truly massive real world datasets, containing tens of millions of time series.     

Hierarchical clustering of time series data with parametric derivative dynamic time warping

Dynamic Time Warping (DTW) is a popular and efficient distance measure used in classification and clustering algorithms applied to time series data. By computing the DTW distance not on raw data but on the time series of the (first, discrete) derivative of the data, we obtain the so-called Derivative Dynamic Time Warping (DDTW) distance measure. DDTW, used alone, is usually inefficient, but there exist datasets on which DDTW gives good results, sometimes much better than DTW. To improve the performance of the two distance measures, we can combine them into a new single (parametric) distance function. The literature contains examples of the combining of DTW and DDTW in algorithms for supervised classification of time series data. In this paper, we demonstrate that combination of DTW and DDTW can also be applied in a method of time series clustering (unsupervised classification). In particular, we focus on a hierarchical clustering (with average linkage) of univariate (one-dimensional) time series data. We construct a new parametric distance function, combining DTW and DDTW, where a single real number parameter controls the contribution of each of the two measures to the total value of the combined distances. The parameter is tuned in the initial phase of the clustering algorithm. Using this technique in clustering methods requires a different approach (to address certain specific problems) than for supervised methods. In the clustering process we use three internal cluster validation measures (measures which do not use labels) and three external cluster validation measures (measures which do use clustering data labels). Internal measures are used to select an optimal value of the parameter of the algorithm, where external measures give information about the overall performance of the new method and enable comparison with other distance functions. Computational experiments are performed on a large real-world data base (UCR Time Series Classification Archive: 84 datasets) from a very broad range of fields, including medicine, finance, multimedia and engineering. The experimental results demonstrate the effectiveness of the proposed approach for hierarchical clustering of time series data. The method with the new parametric distance function outperforms DTW (and DDTW) on the data base used. The results are confirmed by graphical and statistical comparison.     

Dynamic time warping constraint learning for large margin nearest neighbor classification

Nearest neighbor (NN) classifier with dynamic time warping (DTW) is considered to be an effective method for time series classification. The performance of NN-DTW is dependent on the DTW constraints because the NN classifier is sensitive to the used distance function. For time series classification, the global path constraint of DTW is learned for optimization of the alignment of time series by maximizing the nearest neighbor hypothesis margin. In addition, a reduction technique is combined with a search process to condense the prototypes. The approach is implemented and tested on UCR datasets. Experimental results show the effectiveness of the proposed method.     

Faster retrieval with a two-pass dynamic-time-warping lower bound

The dynamic time warping (DTW) is a popular similarity measure between time series. The DTW fails to satisfy the triangle inequality and its computation requires quadratic time. Hence, to find closest neighbors quickly, we use bounding techniques. We can avoid most DTW computations with an inexpensive lower bound (LB_Keogh). We compare LB_Keogh with a tighter lower bound (LB_Improved). We find that LB_Improved-based search is faster. As an example, our approach is 2-3 times faster over random-walk and shape time series.     

Area bound dynamic time warping based fast and accurate person authentication using a biometric pen

The paper presents a modified dynamic time warping (DTW) technique for person authentication based on time series matching obtained from handwriting. The online data has been acquired by a biometric smart pen device. The proposed method allows fast and accurate classification of human individuals based on handwritten PIN words or signature samples. Although classic DTW provides robust distance measurements essential for accurate classification of sequences, it is computationally expensive. To speed up computations we introduce area bound dynamic time warping (AB_DTW) that divides time series into several areas bounded by segments of consecutive zero crossings including local peaks and valleys. Unlike classic DTW which compares whole signals, the proposed AB_DTW warps areas bounded by the local regions. Two kinds of data abstraction formats of area bound—1 dimensional and 2 dimensional—are evaluated. Experimental results show that because of a higher-level data abstraction, the proposed approach is several times faster than classic DTW. Moreover, AB_DTW does not offer substantial loss of accuracy which is required for authentication performance using handwritten PIN words and signatures sampled by biometric pen device.     

基于动态时间弯曲的时序数据聚类算法的研究

时间序列是一类重要的复杂类型数据，时间序列知识发现正成为知识发现的研究热点之一。欧几里的距离及其扩展作为相似测度被广泛应用于时间序列的比较中，但是这种距离测度对数据没有好的鲁棒性。动态时间弯曲技术是基于非线性动态编程的一种模式匹配算法。该文提出了基于动态时间弯曲技术的相似搜索算法，通过计算时序数据之间的最短弯曲路径来获得序列的匹配。对综合控制时序数据进行基于不同距离测度的聚类分析对比结果表明该文提出的算法有很高的精度和对振幅差异、噪声和线性漂移有强的鲁棒性，具有良好的应用价值。     

Rotation-invariant similarity in time series using bag-of-patterns representation

For more than a decade, time series similarity search has been given a great deal of attention by data mining researchers. As a result, many time series representations and distance measures have been proposed. However, most existing work on time series similarity search relies on shape-based similarity matching. While some of the existing approaches work well for short time series data, they typically fail to produce satisfactory results when the sequence is long. For long sequences, it is more appropriate to consider the similarity based on the higher-level structures. In this work, we present a histogram-based representation for time series data, similar to the ??bag of words?? approach that is widely accepted by the text mining and information retrieval communities. We performed extensive experiments and show that our approach outperforms the leading existing methods in clustering, classification, and anomaly detection on dozens of real datasets. We further demonstrate that the representation allows rotation-invariant matching in shape datasets.     

Scalable time series classification

Time series classification tries to mimic the human understanding of similarity. When it comes to long or larger time series datasets, state-of-the-art classifiers reach their limits because of unreasonably high training or testing times. One representative example is the 1-nearest-neighbor dynamic time warping classifier (1-NN DTW) that is commonly used as the benchmark to compare to. It has several shortcomings: it has a quadratic time complexity in the time series length and its accuracy degenerates in the presence of noise. To reduce the computational complexity, early abandoning techniques, cascading lower bounds, or recently, a nearest centroid classifier have been introduced. Still, classification times on datasets of a few thousand time series are in the order of hours. We present our Bag-Of-SFA-Symbols in Vector Space classifier that is accurate, fast and robust to noise. We show that it is significantly more accurate than 1-NN DTW while being multiple orders of magnitude faster. Its low computational complexity combined with its good classification accuracy makes it relevant for use cases like long or large amounts of time series or real-time analytics.     

Exact indexing for massive time series databases under time warping distance

Among many existing distance measures for time series data, Dynamic Time Warping (DTW) distance has been recognized as one of the most accurate and suitable distance measures due to its flexibility in sequence alignment. However, DTW distance calculation is computationally intensive. Especially in very large time series databases, sequential scan through the entire database is definitely impractical, even with random access that exploits some index structures since high dimensionality of time series data incurs extremely high I/O cost. More specifically, a sequential structure consumes high CPU but low I/O costs, while an index structure requires low CPU but high I/O costs. In this work, we therefore propose a novel indexed sequential structure called TWIST (Time Warping in Indexed Sequential sTructure) which benefits from both sequential access and index structure. When a query sequence is issued, TWIST calculates lower bounding distances between a group of candidate sequences and the query sequence, and then identifies the data access order in advance, hence reducing a great number of both sequential and random accesses. Impressively, our indexed sequential structure achieves significant speedup in a querying process. In addition, our method shows superiority over existing rival methods in terms of query processing time, number of page accesses, and storage requirement with no false dismissal guaranteed.     

Template-Based Continuous Speech Recognition

Despite their known weaknesses, hidden Markov models (HMMs) have been the dominant technique for acoustic modeling in speech recognition for over two decades. Still, the advances in the HMM framework have not solved its key problems: it discards information about time dependencies and is prone to overgeneralization. In this paper, we attempt to overcome these problems by relying on straightforward template matching. The basis for the recognizer is the well-known DTW algorithm. However, classical DTW continuous speech recognition results in an explosion of the search space. The traditional top-down search is therefore complemented with a data-driven selection of candidates for DTW alignment. We also extend the DTW framework with a flexible subword unit mechanism and a class sensitive distance measure-two components suggested by state-of-the-art HMM systems. The added flexibility of the unit selection in the template-based framework leads to new approaches to speaker and environment adaptation. The template matching system reaches a performance somewhat worse than the best published HMM results for the Resource Management benchmark, but thanks to complementarity of errors between the HMM and DTW systems, the combination of both leads to a decrease in word error rate with 17% compared to the HMM results     

高效的时间序列下界技术

针对时间序列数据,提出一种新的基于动态时间弯曲的下界技术,该技术首先基于分段聚集近似的线性表示对原始序列进行降维,同时生成查询序列的网格最小边界矩形近似表示,然后利用基于动态时间弯曲距离对两者下界距离度量。实验结果表明,该下界技术与以往相关技术相比,能够产生更大的下界距离,具有更强的紧凑度、裁剪搜索空间能力以及更短的运行时间,有利于时间序列数据挖掘。