Discovering similar multidimensional trajectories

By Michail Vlachos, Dimitrios Gunopulos, George Kollios. Proceedings / International Conference on Data Engineering, 2002

1,667 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Signal Processing

View Paper: https://www.semanticscholar.org/paper/ccd11c8db43da97e6d5a3d0b10e8497eca54597e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Signal Processing

Citations in the corpus

A symbolic representation of time series, with implications for streaming algorithms (2003, methodology): “A Symbolic Representation of Time Series, with Implications for Streaming Algorithms Jessica Lin Eamonn Keogh Stefano Lonardi Bill Chiu University of California -Riverside Computer Science & Engineering Department Riverside, CA 92521, USA {jessica, eamonn, stelo, bill}@cs.ucr.edu ABSTRACT The…”
On The Marriage of Lp-norms and Edit Distance (2004, methodology): “Finally, we used L1-norm, DTW, EDR, ERP and LCSS to rank the five time series relative to Q.”
Plato (2020, neutral): “Following the assumptions in [43, 10, 56] we assume that time is discrete and the resolution of any two time series is the same.”
Deep Representation Learning for Trajectory Similarity Computation (2018, methodology): “similarity has also been recognized by researchers, and a number of classic methods have been proposed, such as dynamic time wrapping (DTW) [6], longest common subsequence (LCSS) [7], edit distance with real penalty (ERP) [8], and edit distance on real sequences (EDR) [9].”
Pattern graph tracking-based stock price prediction using big data (2018, methodology): “There are various methods for pattern matching such as Euclidean distance, Dynamic Time Warping (DTW)([4]), Edit Distance with Real Penalty (ERP)([15]), Longest Common Subsequence (LCSS)([44]) and Edit Distance on Real Sequence (EDR)([16]).”
Trajectory Learning for Activity Understanding: Unsupervised, Multilevel, and Long-Term Adaptive Approach (2011, methodology): “The LCSS distance for trajectories suggested by Vlachos et al. [35] is defined as D LCSS ðF i ; F j Þ ¼ 1 À LCSSðF i ; F j Þ minðT i ; T j Þ ; ð6Þ where the LCSSðF i ; F j Þ value specifies the number of matching points between two trajectories of different length, T i and T j .”
Exact indexing of dynamic time warping (2002, background): “However, there is an increasing awareness that the Euclidean distance is a very brittle distance measure (Aach and Church 2001; Bar-Joseph et al. 2002; Berndt and Clifford 1994; Chu et al. 2002; Diez and Gonzalez 2000; Kadous 1999; Keogh and Pazzani 2000; Kollios et al. 2002; Schmill et al. 1999; Yi”
Weighted dynamic time warping for time series classification (2011, methodology): “The constant e, where 0oeo1, is the matching threshold (please refer to [24] in details).”
Probabilistic discovery of time series motifs (2003, background): “Each time series is normalized to have mean zero and a standard deviation of one before calling the distance function, because it is well understood that it is meaningless to compare time series with different offsets and amplitudes [6, 21, 35, 36].”
Path Similarity Skeleton Graph Matching (2008, methodology): “The Dynamic Time Warping (DTW) distance has been shown to be superior to the Euclidean distance of sequences in many cases [40][ 41 ][42][43].”
Rotation-invariant similarity in time series using bag-of-patterns representation (2012, background): “Other elastic distance measures exist for time series, such as Longest Common SubSequence (LCSS) (Vlachos et al. 2002), Edit distance with Real Penalty (ERP) (Chen and Ng 2004) and Time-Warp Edit Distance (TWED) (Marteau 2009).”
Trajectory clustering: a partition-and-group framework (2007, methodology): “A typical data analysis task is to find objects that have moved in a similar way [20].”
FastDTW: Toward Accurate Dynamic Time Warping in Linear Time and Space (2004, background): “which is a notoriously inaccurate similarity measure for time series [15].”
A Survey on Trajectory Data Management, Analytics, and Learning (2020, methodology): “r each item but exploits a bound to minimize the number of expensive computations applied during processing. This technique is commonly used to accelerate time series similarity search and kNN search [159]. By estimating the upper bound of the similarity score and comparing it to the kth item in a m”
Trajectory Clustering and an Application to Airspace Monitoring (2010, background): “Vlachos et al. used similarity functions based on the longest common subsequence (LCS) to discover similar multidimensional trajectories [8].”