Author Email: email@example.com
Time series is a sequential collection of values with respect to time obtained from various applications. The time series data have basic features like huge data size, high dimensionality with characteristics like trend, cyclical, seasonal, and irregular. The cumulative use of time series data has initiated a great deal of research and development attempts in the field of data mining. Dimensionality of time series is directly proportional to the efficiency of various data mining algorithms used for time series analysis. In this paper, a widespread review on the existing time series dimensionality reduction methods is given. The chief objective of this paper is to aid interested researchers to have a general idea about the current investigation in time series dimensionality reduction methods and identify their potential research direction to advance investigation in the same. The papers also discuss about the possibilities of using automata model for time series dimensionality reduction.
Dimensionality, Time Series, Automata, Data Mining
Large amount of data is being produced by various organizations in the world; the inventions of social networking websites are continuously adding to the data repositories all over the world. Most of this data is time dependent i.e. the time series data and hence has some features like cyclical, seasonal, trend, irregular. Storing this huge data needs lot of memory moreover analysis of this data for extracting these features from it consumes lot of time. Thus, to reduce the time required for extracting these information the given time series has to be represented in lower dimension thus storing only necessary values that depicts some reasonable information and helps in important data mining tasks like classification, clustering, forecasting, etc. by feature extraction, matching and computation of parametric values of the time series.
Existing methods used for dimensionality reduction in time series are sampling, Piecewise Aggregate Approximation (PAA), Dynamic Time Warping (DWT), Piecewise Linear Approximation (PLA), Piecewise Trend approximation (PTA), Piecewise Cloud Approximation (PWCA), Symbolic Aggregate Approximation (SAX), .
Even though by applying any of the above method on time series it is possible to extract required information form time series, still there is scope of improvement in the efficiency of time series mining tasks by designing new methods for time series dimensionality reduction as no exact method is best it depends on application domain.
480 total views, 1 views today