A Faster Reduction of the Dynamic Time Warping Distance to the Longest Increasing Subsequence Length

被引:3
作者
Sakai, Yoshifumi [1 ]
Inenaga, Shunsuke [2 ,3 ]
机构
[1] Tohoku Univ, Grad Sch Agr Sci, Sendai, Miyagi, Japan
[2] Kyushu Univ, Dept Informat, Fukuoka, Japan
[3] Japan Sci & Technol Agcy, PRESTO, Kawaguchi, Saitama, Japan
关键词
String algorithms; Dynamic time warping distance; Longest increasing subsequence; Semi-local sequence comparison; ALGORITHM;
D O I
10.1007/s00453-022-00968-2
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
The similarity between a pair of time series, i.e., sequences of indexed values in time order, is often estimated by the dynamic time warping (DTW) distance, instead of any in the well-studied family of measures including the longest common subsequence (LCS) length and the edit distance. Although it may seem as if the DTW and the LCS(-like) measures are essentially different, we reveal that the DTW distance can be represented by the longest increasing subsequence (LIS) length of a sequence of integers, which is the LCS length between the integer sequence and itself sorted. For a given pair of time series of length n such that the dissimilarity between any elements is an integer between zero and c, we propose an integer sequence that represents any substring-substring DTW distance as its band-substring LIS length. The length of the produced integer sequence is O(cn(2)), which can be translated to O(n(2)) for constant dissimilarity functions. To demonstrate that techniques developed under the LCS(like) measures are directly applicable to analysis of time series via our reduction of DTWto LIS, we present time-efficient algorithms for DTW-related problems utilizing the semi-local sequence comparison technique developed for LCS-related problems.
引用
收藏
页码:2581 / 2596
页数:16
相关论文
共 31 条
  • [1] Tight Hardness Results for LCS and other Sequence Similarity Measures
    Abboud, Amir
    Backurs, Arturs
    Williams, Virginia Vassilevska
    [J]. 2015 IEEE 56TH ANNUAL SYMPOSIUM ON FOUNDATIONS OF COMPUTER SCIENCE, 2015, : 59 - 78
  • [2] Quadratic Conditional Lower Bounds for String Problems and Dynamic TimeWarping
    Bringmann, Karl
    Kuennemann, Marvin
    [J]. 2015 IEEE 56TH ANNUAL SYMPOSIUM ON FOUNDATIONS OF COMPUTER SCIENCE, 2015, : 79 - 97
  • [4] Fast computation of a longest increasing subsequence and application
    Crochemore, Maxime
    Porat, Ely
    [J]. INFORMATION AND COMPUTATION, 2010, 208 (09) : 1054 - 1059
  • [5] Coarse-DTW for Sparse Time Series Alignment
    Dupont, Marc
    Marteau, Pierre-Francois
    [J]. ADVANCED ANALYSIS AND LEARNING ON TEMPORAL DATA, AALTD 2015, 2016, 9785 : 157 - 172
  • [6] Froese V., 2020, ABS190303003 CORR
  • [7] Gold O., 2020, ABS160705994V4 CORR
  • [8] Dynamic Time Warping and Geometric Edit Distance: Breaking the Quadratic Barrier
    Gold, Omer
    Sharir, Micha
    [J]. ACM TRANSACTIONS ON ALGORITHMS, 2018, 14 (04)
  • [9] LINEAR SPACE ALGORITHM FOR COMPUTING MAXIMAL COMMON SUBSEQUENCES
    HIRSCHBERG, DS
    [J]. COMMUNICATIONS OF THE ACM, 1975, 18 (06) : 341 - 343
  • [10] Hwang Y., 2019, MLDM 2019, P748