Distance measures for effective clustering of ARIMA time-series

By Konstantinos Kalpakis, D. Gada, V. Puttagunta. Proceedings 2001 IEEE International Conference on Data Mining, 2001

498 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Signal Processing

View Paper: https://www.semanticscholar.org/paper/cfea224977f062a1f279a48e310018f396f0746a

How researchers see this paper

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

In Economics and Econometrics

In Signal Processing

Citations in the corpus

Data Mining for the Internet of Things: Literature Review and Challenges (2015, methodology): “Important research works include ARMA [84] and the time series bitmaps research [85].”
Time-Series Classification with COTE: The Collective of Transformation-Based Ensembles (2015, methodology): “The most common way to use the ACF in time-series data mining is to ﬁt an AR model (i.e. use W with p set to minimize AIC) to each series, then base similarity on differences in model parameters, using, for example, Euclidean distance (see, e.g., [39], [40], [41]).”
Time-series data mining (2012, methodology): “Several parametric temporal models may be considered, including statistical modeling by feature extraction [Nanopoulos et al. 2001], ARMA models [Kalpakis et al. 2001], Markov Chains (MCs) [Sebastiani et al. 1999], or HMM [Panuccio et al. 2002].”
A Unified Framework for Model-based Clustering (2003, background): “For example, a wide variety of distance measures have been proposed for clustering sequences (Geva and Kerem, 1998; Kalpakis et al., 2001; Qian et al., 2001).”
k-Shape: Efficient and Accurate Clustering of Time Series (2016, background): “The ubiquity of time series has generated a substantial interest in querying [2, 38, 39, 41, 52, 61, 65], indexing [8, 11, 34, 35, 37, 63], classification [30, 47, 58, 70], clustering [36, 45, 54, 69, 71], and modeling [3, 31, 68] of such data.”
Metabolomics and Multi-Omics Integration: A Survey of Computational Methods and Resources (2020, background): “Models based on the assumed statistical properties of the processes which create the time series have also been explored, such as auto-regressive moving average (ARMA) models [83,84] or hidden Markov models (HMM) [85,86].”
Time series feature extraction for data mining using DWT and DFT (2003, methodology): “…literature and other sophisticated distance functions have been combined with clustering algorithms, for example k-Means clustering [13], [15] (with ARIMA based distance), [20] (with DWT), hierarchical clustering [30] (with Euclidean distance), [27] (with dynamic time warping distance),…”
Time-focused clustering of trajectories of moving objects (2006, background): “…subsequence of two series (e.g., see Vlachos, Gunopulos, & Kollios, 2002), the count of common subsequences of length two (Agrawal, Lin, Sawhney, & Shim, 1995), the domain-dependent extraction of a single representative value for the whole series Kalpakis, Gada, and Puttagunta (2001), and so on.”
A symbolic representation of time series, with implications for streaming algorithms (2003, methodology): “• Clustering: Find natural groupings of the time series in database DB under some similarity/dissimilarity measure D(Q,C) [21,25].”
k-Shape: Efficient and Accurate Clustering of Time Series (2015, methodology): “algorithms, statistical-based approaches use measures that summarize characteristics [82], coefficients of models [38], or portions of time series (i.”
Towards parameter-free data mining (2004, methodology): “The LPC Cepstra approach of [17] and the similar Autocorrelation method of [35] both had Q = 0.”
Characteristic-Based Clustering for Time Series Data (2006, background): “The mining of time series data has attracted great attention in the data mining community in recent years (Bradley and Fayyad, 1998; Halkidi et al., 2001; Kalpakis et al., 2001; Keogh et al., 2003; Wang and Wang, 2000).”
Finding Motifs in Time Series (2002, background): “, query by content) has received much attention and may now be essentially regarded as a solved problem [1, 8, 13, 21, 22, 23, 35, 40].”
A review on time series data mining (2011, methodology): “Kalpakis et al. (2001) propose to cluster ARIMA time series, using the partition around the medoids method.”
Experiencing SAX: a novel symbolic representation of time series (2007, methodology): “In particular, we will demonstrate the utility of our representation on the classic data mining tasks of clustering (Kalpakis et al. 2001), classification (Geurts 2001), indexing (Agrawal et al.”
Big IoT Data Analytics: Architecture, Opportunities, and Open Research Challenges (2017, background): “Research related to time series representation includes ARMA [62], bitmaps [63], and wavelet functions [64].”