Adaptive piecewise and symbolic aggregate approximation as an improved representation method for heat waves detection

摘要: Mining time series has attracted an increasing interest due to its wide applications in finance, industry, biology, environment, and so on. In order reduce execution storage space, many high level representations or abstractions of the raw data have been proposed including Discrete Fourier Transform (DFT), Wavelet (DWT), Piecewise Aggregate Approximation (PAA) Symbolic approXimation (SAX). this paper, we introduce a novel adaptive piecewise symbolic aggregate approximation (APAA/ASAX) which creates segments variable length automatically adapt any segment local condition variability difference average value current values is defined. The each from APAA represented as symbol ordered alphabet generating modified version for SAX called (ASAX). This straightforwardly allows handle more versatile definition event duration. method APAA/ASAX was used locating heat waves patterns real-world datasets daily temperature information, year 1970 until 2009. experimental results show that representation able locate heatwave events huge databases. Advantages regarding traditional PAA are mainly based on being constrain-free fixed schemes length. It also highlights ability self-tuning depending characteristics. means flat proposes lower number dimensionality than case deal with variability. approach will be use those looking extreme series.