13-10-2010, 10:15 AM
Abstract
Prediction of an event at a time series is quite important for engineering and economy problems. Time series data mining combines the fields of time series analysis and data mining techniques. This method creates a set of methods that reveal hidden temporal patterns that are characteristic and predictive of time series events. Time series data mining examines the time series in a phase space. In this paper, a prediction algorithm using time series data mining based on fuzzy logic is proposed. Earthquake prediction has been done from a synthetic earthquake time series by using investigating method at first step ago. Time series has been transformed to phase space by using nonlinear time series analysis and then fuzzy logic has been used to prediction optimal values of important parameters characterizing the time series events. Truth of prediction algorithm based fuzzy logic has been proved by application results.
For more details, please visit
http://wasetjournals/waset/v51/v51-17.pdf