基于趋势转折点的时间序列分段线性表示

分段线性表示是时间序列降维的有效方法。在总结分析序列趋势变化特点的基础上，提出了一种基于趋势转折点的时间序列分段线性表示算法。首先定义了趋势转折点作为时间序列分段点的备选集，以点到区域的距离度量趋势转折点的重要性，再根据给定的阈值选择重要趋势转折点作为分段点，对时间序列进行分段线性表示。通过与其他6种方法进行实验比较，结果表明：所提方法在具有较好的拟合质量和适应能力以及对转折点明显的序列，都表现出较强的抗噪声干扰能力。

Time series piecewise linear representation based on trend transition point

Piecewise Linear Representation（PLR） is an effective method to reduce the dimension of time series.A method of time series PLR algorithm is put forward based on the summarization of the characteristic of the series trends diversification.Firstly,this paper defines the trend transition points for the candidacy sets of time series＇segmenting points,and uses the distances between point and area to measure the weightiness of these points,then regards the important trend transition points that are chose according to specified threshold as the segmenting point for the PLR of time series.Compared with other six segmentation algorithms,the experiments show that the proposed algorithm appears better fitting quality and suitability.Especially for the time series which have the distinct transition point,the proposed algorithm has better ability of anti-noise-jamming.