scholarly journals A New Method for Piecewise Linear Representation of Time Series Data

2012 ◽  
Vol 25 ◽  
pp. 1097-1103 ◽  
Author(s):  
Jiajie Zhou ◽  
Gang Ye ◽  
Dan Yu
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Piecewise Linear ◽  
Linear Representation ◽  
New Method ◽  
Series Data ◽  
Piecewise Linear Representation ◽  
Representation Of Time

Time Series Data Mining in Process Industry

2012 ◽  
Vol 532-533 ◽  
pp. 1069-1074
Author(s):  
Jia Ren ◽  
Jin Feng Gao
Keyword(s):  
Data Mining ◽  
Time Series ◽  
Large Scale ◽  
Time Series Data ◽  
Piecewise Linear ◽  
Linear Representation ◽  
Series Data ◽  
Time Series Data Mining ◽  
Piecewise Linear Representation ◽  
Mining Procedure

time series; data mining; knowledge discovery; trend extremum representation. Abstract. In recent years, there has been an explosion of interest in mining time series databases. In this paper, we make some attempts to mining process industrial time series. As with most computer science problems, representation of the data is the key to efficient and effective solutions. We introduce a novel algorithm, Trend Extremum Representation, which is empirically proved to be superior to Piecewise Linear Representation and Important Points Representation in manipulating large-scale industrial data. Then, subsequent mining procedure is undertaken. Through clustering analysis and association rule discovery, several useful rules are derived for differentiating normal and abnormal events in everyday operations.

A New Method of Time Series Piecewise Linear Representation

2013 ◽  
Vol 705 ◽  
pp. 569-573
Author(s):  
Qiang Wang
Keyword(s):  
Time Series ◽  
Piecewise Linear ◽  
Linear Representation ◽  
New Method ◽  
Time Data ◽  
Hyperbolic Tangent ◽  
Piecewise Linear Representation ◽  
Tangent Function ◽  
Dynamic Growth ◽  
Over Time

To meet requirements of time data dynamic growth , and reflect the different effect to the different segment of time series over time, a new method of piecewise linear representation, called tangent piecewise aggregate approximation (TPAA) is proposed based on hyperbolic tangent function. The method can not only meet requirements of time data dynamic growth, but also reflect time property of the time series. Compared with the existing methods, TPAA method can effectively query time series online.

Piecewise linear representation of time series based on mean trend in sliding window

2011 ◽  
Vol 15 (5) ◽  
pp. 473-478
Author(s):  
Tong-yu Yuan ◽  
Shao-chun Wu ◽  
Jian Zhang ◽  
Rong-rong Gu ◽  
Gao-zhao Chen ◽  
...  
Keyword(s):  
Time Series ◽  
Piecewise Linear ◽  
Sliding Window ◽  
Linear Representation ◽  
Piecewise Linear Representation ◽  
Mean Trend ◽  
Representation Of Time

Earthquake Prediction Research Based on the Mining in Time Series of Groundwater Temperature

2012 ◽  
Vol 532-533 ◽  
pp. 1016-1020 ◽  
Author(s):  
Jian Zhang ◽  
Yi Lin Lu ◽  
Shao Chun Wu
Keyword(s):  
Time Series ◽  
Earthquake Prediction ◽  
Time Series Data ◽  
Sequence Data ◽  
Piecewise Linear ◽  
Extreme Points ◽  
Linear Representation ◽  
Series Data ◽  
Groundwater Temperature ◽  
Symbol Sequence

Earthquake prediction has always been an extremely important and difficult research topic. A road map was proposed in this paper to capture useful information for earthquake prediction by exploring the time sequence data of groundwater temperature. Firstly, the triangle extreme points and the trend turning points are employed for the piecewise linear representation of the time series data. Then the segmentation is classified and symbolized by slope, and symbol sequence is simplified further according to the simplification rules. Finally, the earthquake catalogue data and the symbol sequence are jointly preprocessed with a new method to form transaction-like data, which then be treated by association analysis to extract earthquake prediction knowledge. The results of experiment show that this processing flow is an effective way to provide valuable information about earthquake prediction.

scholarly journals A New Method to Detect Outliers in High-frequency Time Series

2018 ◽  
Vol 8 (1) ◽  
pp. 16
Author(s):  
Ilaria Lucrezia Amerise ◽  
Agostino Tarsitano
Keyword(s):  
Time Series ◽  
High Frequency ◽  
Time Series Data ◽  
New Method ◽  
Series Data ◽  
Absolute Difference ◽  
Data Preparation ◽  
Simple Method ◽  
Pass Through ◽  
Nonparametric Procedure

The objective of this research is to develop a fast, simple method for detecting and replacing extreme spikes in high-frequency time series data. The method primarily consists  of a nonparametric procedure that pursues a balance between fidelity to observed data and smoothness. Furthermore, through examination of the absolute difference between original and smoothed values, the technique is also able to detect and, where necessary, replace outliers with less extreme data. Unlike other filtering procedures found in the literature, our method does not require a model to be specified for the data. Additionally, the filter makes only a single pass through the time series. Experiments  show that the new method can be validly used as a data preparation tool to ensure that time series modeling is supported by clean data, particularly in a complex context such as one with high-frequency data.

Sign in / Sign up

Export Citation Format

Share Document