Similarity search over time-series data using wavelets

By Ivan Popivanov, Renée J. Miller. Proceedings / International Conference on Data Engineering, 2002

320 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/6d57e29ddb68c91256e59d82e8afe321152aa357

How researchers see this paper

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

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Indexing spatio-temporal trajectories with Chebyshev polynomials (2004, methodology): “Unlike some other frameworks, like wavelet decompositions [4, 27, 9, 18], we do not require the power-of-2 assumption.”
Using discrete wavelet transforms to analyze trends in streamflow and precipitation in Quebec and Ontario (1954–2008) (2012, methodology): “Daubechies wavelets provide compact support (Vonesch et al., 2007), indicating that the wavelets have non-zero basis functions over a finite interval, as well as full scaling and translational orthonormality properties (Popivanov and Miller, 2002; de Artigas et al., 2006).”
Evaluating a coupled discrete wavelet transform and support vector regression for daily and monthly streamflow forecasting (2014, methodology): “In this study, a set of Daubechies (db) wavelets were considered, because they have certain features that are very important for localizing events in the time-dependent signals (Daubechies, 1990; Popivanov and Miller, 2002).”
On the Need for Time Series Data Mining Benchmarks: A Survey and Empirical Demonstration (2002, background): “In contrast if we worked with the Network, ERPdata and Fetal EEG datasets we could conclude that there is no real difference between DFT and Haar, as suggested by Figure 3.”
StatStream: Statistical Monitoring of Thousands of Data Streams in Real Time (2002, methodology): “Other techniques such as Discrete Wavelet Transform (DWT) [5, 26, 22, 12], Singular Value Decomposition (SVD)[18] and Piecewise Constant Approximation (PCA)[27, 17] are also proposed for similarity search.”
Time series feature extraction for data mining using DWT and DFT (2003, background): “[22] pursued the suggestion by Chan [6], that wavelet bases other than Haar, might be better for certain data sets.”
Characteristic-Based Clustering for Time Series Data (2006, background): “• Distance Measures: Euclidean distance, the most commonly used metric (Agrawal et al., 1993; Chan and Fu, 1999; Chu and Wong, 1999; Faloutsos et al., 1994; Keogh et al., 2001; Popivanov and Miller, 2002), is certainly not the only metric available for measuring the similarity between data series.”
Warping indexes with envelope transforms for query by humming (2003, methodology): “In addition to DFT [1, 25, 35], Discrete Wavelet Transform (DWT) [6, 30, 23], Singular Value Decomposition (SVD)[16], Piecewise Aggregate Approximation (PAA)[32, 13] and Adaptive Piecewise Constant Approximation [14] approaches have also been proposed for similarity searching.”
Time-series data mining (2012, methodology): “A large number of wavelet functions have been used in the literature like Haar [Chan et al. 2003], Daubechies [Popivanov and Miller 2002], or Coi.ets [Shasha and Zhu 2004].”
Time-series clustering - A decade review (2015, background): “Accordingly, many researches had been carried out focusing on representation and dimensionality reduction [84–90].”
TSclust: An R Package for Time Series Clustering (2014, background): “…series analysis can be seen in Percival and Walden (2006) and some interesting references where wavelets are used to clustering time series are Struzik and Siebes (1999), Chan and Fu (1999), Popivanov and Miller (2002), Chan, Fu, and Yu (2003) and Zhang, Ho, Zhang, and Lin (2006), among others.”
A review on time series data mining (2011, background): “Popivanov and Miller (2002) show that a large class of wavelet transformations can be used for time series representation.”
Querying and mining of time series data: experimental comparison of representations and distance measures (2008, result): “For example, one paper claims " wavelets outperform the DFT " [34], another claims " DFT filtering performance is superior to DWT " [19] and yet another claims " DFT-based and DWT-based techniques yield comparable results " [41].”
On The Marriage of Lp-norms and Edit Distance (2004, methodology): “Examples of dimensionality reduction techniques include Single Value Decomposition [19], Discrete Wavelet Transform [20] [22], Piecewise Aggregate Approximation [15] [29], and Adaptive Piecewise Constant Approximation [14].”
Experimental comparison of representation methods and distance measures for time series data (2010, background): “Many techniques have been proposed for representing time series with reduced dimensionality, for example: Discrete Fourier Transformation (DFT) (Faloutsos et al. 1994), Single Value Decomposition (SVD) (Faloutsos et al. 1994), Discrete Cosine Transformation (DCT) (Korn et al. 1997), Discrete Wavelet”