Fuzzy-Based Shapelets for Mining Climate Change Time Series Patterns

It is difficult to identify visualized multi-climate change patterns from time series data due to the fact that the data begin to look similar over time. Traditionally, time series weather patterns are presented in the form of a linear graph, which is limited to discovering understandable climate change patterns. On the other hand, the Symbolic Aggregate Approximation (SAX) algorithm based on the Piecewise Aggregate Approximation (PAA), which is known as a popular method to solve this problem, has its limitations. Therefore, the aim of this research was to propose a fuzzy-based symbolic data representation, known as a Shapelet Patterns Algorithm (SPA), in order to come up with a Shapelet Pattern (SP) for climate change. The shapelet pattern was able to visualize climate change patterns in the form of coloured shapes to indicate annual changes in temperature patterns, such as cool, warm, hot and very hot. The experiment used the climate change data for 1985-2008 gathered from the Petaling Jaya station in the state of Selangor, Malaysia. The shapelet patterns revealed seven types of climate change patterns and presented detailed information on climate changes that can aid climate change experts in better decision making.

[1]  Siegfried Gottwald,et al.  Universes of Fuzzy Sets and Axiomatizations of Fuzzy Set Theory. Part II: Category Theoretic Approaches , 2006, Stud Logica.

[2]  Witold Pedrycz,et al.  Fuzzy set technology in knowledge discovery , 1998, Fuzzy Sets Syst..

[3]  Tak-Chung Fu,et al.  A review on time series data mining , 2011, Eng. Appl. Artif. Intell..

[4]  Tor Arne Johansen,et al.  Fuzzy model based control: stability, robustness, and performance issues , 1994, IEEE Trans. Fuzzy Syst..

[5]  Nitin K. Tripathi,et al.  An artificial neural network model for rainfall forecasting in Bangkok, Thailand , 2008 .

[6]  Thomas Nocke,et al.  Methods for the visualization of clustered climate data , 2004, Comput. Stat..

[7]  W. Pedrycz,et al.  Fuzzy computing for data mining , 1999, Proc. IEEE.

[8]  F. Morabito,et al.  Fuzzy time series approach for disruption prediction in Tokamak reactors , 2003 .

[9]  Eamonn J. Keogh,et al.  Logical-shapelets: an expressive primitive for time series classification , 2011, KDD.

[10]  Azuraliza Abu Bakar,et al.  Harmony Search algorithm for optimal word size in symbolic time series representation , 2011, 2011 3rd Conference on Data Mining and Optimization (DMO).

[11]  Eamonn J. Keogh,et al.  A symbolic representation of time series, with implications for streaming algorithms , 2003, DMKD '03.

[12]  S. Sivanandam,et al.  Introduction to Fuzzy Logic using MATLAB , 2006 .

[13]  Robert S. Laramee,et al.  The State of the Art in Flow Visualization: Dense and Texture‐Based Techniques , 2004, Comput. Graph. Forum.

[14]  Ken Brodlie,et al.  Recent Advances in Volume Visualization , 2001, Comput. Graph. Forum.

[15]  Vincent R. Gray.  Climate Change 2007: The Physical Science Basis Summary for Policymakers , 2007 .

[16]  Eamonn J. Keogh,et al.  Time series shapelets: a new primitive for data mining , 2009, KDD.

[17]  Ajith Abraham,et al.  Adaptation of Fuzzy Inference System Using Neural Learning , 2005 .

[18]  Ronald R. Yager.  Database discovery using fuzzy sets , 1996 .

[19]  P. C. Nayak,et al.  Fuzzy computing based rainfall–runoff model for real time flood forecasting , 2005 .

[20]  James C. Bezdek.  The thirsty traveler visits Gamont: a rejoinder to "Comments on fuzzy sets-what are they and why?" , 1994, IEEE Trans. Fuzzy Syst..

[21]  George Yannis,et al.  On statistical inference in time series analysis of the evolution of road safety. , 2013, Accident; analysis and prevention.

[22]  Eamonn J. Keogh,et al.  iSAX: indexing and mining terabyte sized time series , 2008, KDD.

[23]  C. Ahrens,et al.  Meteorology Today: An Introduction to Weather, Climate, and the Environment , 1982 .

[24]  Jiawei Han,et al.  Data Mining: Concepts and Techniques , 2000 .

[25]  Dimitrios Gunopulos,et al.  Mining Time Series Data , 2005, Data Mining and Knowledge Discovery Handbook.

[26]  Gwo-Fong Lin,et al.  Time series forecasting by combining the radial basis function network and the self‐organizing map , 2005 .

[27]  Stephen Shaoyi Liao,et al.  Finding relevant sequences in time series containing crisp, interval, and fuzzy interval data , 2004, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[28]  Nikola K. Kasabov,et al.  DENFIS: dynamic evolving neural-fuzzy inference system and its application for time-series prediction , 2002, IEEE Trans. Fuzzy Syst..