Time series symbolization based on singular event feature clustering
QU Wen-long, Kejun Zhang · Systems engineering and electronics · 2006
A feature extraction method of singular event based on multi-scale wavelet is presented.Time series is divided into event subsequences with independent trend.The event subsequences represented by transformed parameters are clustered by a hybrid genetic algorithm to implement symbolization.Each event is identified by the cluster it belongs to.The proposed method is applied to unsteady financial time series symbolization.Experimental result shows that the method is effective.