摘要
针对时间序列的数据挖掘首先需要将时间序列(Time Series)数据转换为离散的符号序列(Symbol Sequence)。在前人的基础上,将界标模型和分段线性化进行了结合,以关键点作为分段依据,以最大似然函数和最小二乘法来拟合各分段线性拟合函数;此方法的优点在于符合人体生理实验结果,考虑了时间序列中的噪声。
For the first time series data mining needs of the time series data into discrete sequence of symbols,in the previous article,based on the landmark model and the linear segments were combined to key point as the segment based on the maximum likelihood function and the least squares method to fit the piecewise linear function with the advantage in line with physiological results,considering the noise in time sequence.
出处
《软件导刊》
2011年第10期142-144,共3页
Software Guide
