A time series clustering method based on hypergraph partitioning

2016 ◽  
Author(s):  
Zhibo Zhu ◽  
Qinke Peng ◽  
Xinyu Guan
Keyword(s):  
Time Series ◽  
Clustering Method ◽  
Time Series Clustering ◽  
Hypergraph Partitioning

scholarly journals CLUSTERING OF RAINFALL DISTRIBUTION PATTERNS IN PENINSULAR MALAYSIA USING TIME SERIES CLUSTERING METHOD

2019 ◽  
Vol 38 ◽  
pp. 84-99
Author(s):  
Noratiqah Mohd Ariff ◽  
Mohd Aftar Abu Bakar ◽  
Sharifah Faridah Syed Mahbar ◽  
Mohd Shahrul Mohd Nadzir
Keyword(s):  
Time Series ◽  
Distribution Patterns ◽  
Peninsular Malaysia ◽  
Clustering Method ◽  
Rainfall Distribution ◽  
Time Series Clustering

scholarly journals Trend analysis using non-stationary time series clustering based on the finite element method

2014 ◽  
Vol 21 (3) ◽  
pp. 605-615 ◽  
Author(s):  
M. Gorji Sefidmazgi ◽  
M. Sayemuzzaman ◽  
A. Homaifar ◽  
M. K. Jha ◽  
S. Liess
Keyword(s):  
Time Series ◽  
Temperature Time Series ◽  
Stationary Time Series ◽  
Change Points ◽  
The Finite Element Method ◽  
Clustering Method ◽  
Time Series Clustering ◽  
Stationary Time ◽  
Climatic Time Series ◽  
Temperature Time

Abstract. In order to analyze low-frequency variability of climate, it is useful to model the climatic time series with multiple linear trends and locate the times of significant changes. In this paper, we have used non-stationary time series clustering to find change points in the trends. Clustering in a multi-dimensional non-stationary time series is challenging, since the problem is mathematically ill-posed. Clustering based on the finite element method (FEM) is one of the methods that can analyze multidimensional time series. One important attribute of this method is that it is not dependent on any statistical assumption and does not need local stationarity in the time series. In this paper, it is shown how the FEM-clustering method can be used to locate change points in the trend of temperature time series from in situ observations. This method is applied to the temperature time series of North Carolina (NC) and the results represent region-specific climate variability despite higher frequency harmonics in climatic time series. Next, we investigated the relationship between the climatic indices with the clusters/trends detected based on this clustering method. It appears that the natural variability of climate change in NC during 1950–2009 can be explained mostly by AMO and solar activity.

scholarly journals Simulator Pre-Screening of Underprepared Drivers Prior to Licensing On-Road Examination: Clustering of Virtual Driving Test Time Series Data (Preprint)

2019 ◽  
Author(s):  
David Grethlein ◽  
Flaura Koplin Winston ◽  
Elizabeth Walshe ◽  
Sean Tanner ◽  
Venk Kandadai ◽  
...  
Keyword(s):  
Time Series ◽  
Risk Ratio ◽  
Standard Method ◽  
Domain Knowledge ◽  
Time Series Data ◽  
Series Data ◽  
Test Time ◽  
False Alarms ◽  
Clustering Method ◽  
Time Series Clustering

BACKGROUND A large Midwestern state commissioned a virtual driving test (VDT) to assess driving skills preparedness before the on-road examination (ORE). Since July 2017, a pilot deployment of the VDT in state licensing centers (VDT pilot) has collected both VDT and ORE data from new license applicants with the aim of creating a scoring algorithm that could predict those who were underprepared. OBJECTIVE Leveraging data collected from the VDT pilot, this study aimed to develop and conduct an initial evaluation of a novel machine learning (ML)–based classifier using limited domain knowledge and minimal feature engineering to reliably predict applicant pass/fail on the ORE. Such methods, if proven useful, could be applicable to the classification of other time series data collected within medical and other settings. METHODS We analyzed an initial dataset that comprised 4308 drivers who completed both the VDT and the ORE, in which 1096 (25.4%) drivers went on to fail the ORE. We studied 2 different approaches to constructing feature sets to use as input to ML algorithms: the standard method of reducing the time series data to a set of manually defined variables that summarize driving behavior and a novel approach using time series clustering. We then fed these representations into different ML algorithms to compare their ability to predict a driver’s ORE outcome (pass/fail). RESULTS The new method using time series clustering performed similarly compared with the standard method in terms of overall accuracy for predicting pass or fail outcome (76.1% vs 76.2%) and area under the curve (0.656 vs 0.682). However, the time series clustering slightly outperformed the standard method in differentially predicting failure on the ORE. The novel clustering method yielded a risk ratio for failure of 3.07 (95% CI 2.75-3.43), whereas the standard variables method yielded a risk ratio for failure of 2.68 (95% CI 2.41-2.99). In addition, the time series clustering method with logistic regression produced the lowest ratio of false alarms (those who were predicted to fail but went on to pass the ORE; 27.2%). CONCLUSIONS Our results provide initial evidence that the clustering method is useful for feature construction in classification tasks involving time series data when resources are limited to create multiple, domain-relevant variables.

scholarly journals Simulator Pre-Screening of Underprepared Drivers Prior to Licensing On-Road Examination: Clustering of Virtual Driving Test Time Series Data

10.2196/13995 ◽  
2020 ◽  
Vol 22 (6) ◽  
pp. e13995
Author(s):  
David Grethlein ◽  
Flaura Koplin Winston ◽  
Elizabeth Walshe ◽  
Sean Tanner ◽  
Venk Kandadai ◽  
...  
Keyword(s):  
Time Series ◽  
Risk Ratio ◽  
Standard Method ◽  
Domain Knowledge ◽  
Time Series Data ◽  
Series Data ◽  
Test Time ◽  
False Alarms ◽  
Clustering Method ◽  
Time Series Clustering

Background A large Midwestern state commissioned a virtual driving test (VDT) to assess driving skills preparedness before the on-road examination (ORE). Since July 2017, a pilot deployment of the VDT in state licensing centers (VDT pilot) has collected both VDT and ORE data from new license applicants with the aim of creating a scoring algorithm that could predict those who were underprepared. Objective Leveraging data collected from the VDT pilot, this study aimed to develop and conduct an initial evaluation of a novel machine learning (ML)–based classifier using limited domain knowledge and minimal feature engineering to reliably predict applicant pass/fail on the ORE. Such methods, if proven useful, could be applicable to the classification of other time series data collected within medical and other settings. Methods We analyzed an initial dataset that comprised 4308 drivers who completed both the VDT and the ORE, in which 1096 (25.4%) drivers went on to fail the ORE. We studied 2 different approaches to constructing feature sets to use as input to ML algorithms: the standard method of reducing the time series data to a set of manually defined variables that summarize driving behavior and a novel approach using time series clustering. We then fed these representations into different ML algorithms to compare their ability to predict a driver’s ORE outcome (pass/fail). Results The new method using time series clustering performed similarly compared with the standard method in terms of overall accuracy for predicting pass or fail outcome (76.1% vs 76.2%) and area under the curve (0.656 vs 0.682). However, the time series clustering slightly outperformed the standard method in differentially predicting failure on the ORE. The novel clustering method yielded a risk ratio for failure of 3.07 (95% CI 2.75-3.43), whereas the standard variables method yielded a risk ratio for failure of 2.68 (95% CI 2.41-2.99). In addition, the time series clustering method with logistic regression produced the lowest ratio of false alarms (those who were predicted to fail but went on to pass the ORE; 27.2%). Conclusions Our results provide initial evidence that the clustering method is useful for feature construction in classification tasks involving time series data when resources are limited to create multiple, domain-relevant variables.

Sign in / Sign up

Export Citation Format

Share Document