Big Data Time Series Stream Data Segmentation Methods

Big Data Time Series Stream Data Segmentation Methods

Dima Alberg (Shamoon College of Engineering (SCE), Israel)
DOI: 10.4018/978-1-5225-7598-6.ch004
OnDemand PDF Download:
No Current Special Offers


In this chapter, the authors introduce the interval sliding window (ISW) and interval sliding window and bottom up (ISWAB) algorithms, which are applicable to big data numerical time series data streams and use as input the confidence level parameter rather than the maximum error threshold. The proposed algorithms have two advantages: first, they allow performance comparisons across different time series data streams without changing the algorithm settings, and second, they do not require preprocessing the original time series data stream in order to determine heuristically the reasonable error value. These improvements are very efficient and important in context of big data tine series data streams processing. Finally, an empirical evaluation was performed on two types of time series data.
Chapter Preview


Several high level representations of time series have been proposed in the research literature, including Fourier Transforms (Keogh et al., 2000), Wavelets (Chan & Fu, 1999), Symbolic Mappings (Das, Lin, Mannila, Renganathan, & Smyth, 1998; Perng et al., 2000) and Piecewise Linear Approximation or PLA: (Chan & Fu, 1999; Ge & Smyth, 1999; Hunter & McIntosh, 1998; Junker, Amft, Lukowicz, & Tröster, 2008; Keogh et al., 2004; Lavrenko, Schmill, Lawrie, Ogilvie, Jensen, & Allan, 2000; Li, Yu, & Castelli, 1998; Osaki, Shimada, & Uehara, 1999; Park, Lee, & Chu, 1999; Qu, Wang, & Wang, 1998; Shatkay & Zdonik, 1996; Vullings, Verhaegen, & Verbruggen, 1997; Wang & Wang, 2000).

Complete Chapter List

Search this Book: