What is a time series dataset?
Time series data is data that is collected at different points in time. This is opposed to cross-sectional data which observes individuals, companies, etc. at a single point in time. Because data points in time series are collected at adjacent time periods there is potential for correlation between observations.
How do you classify data as a time series?
A Brief Survey of Time Series Classification Algorithms
- Distance-based (KNN with dynamic time warping)
- Interval-based (TimeSeriesForest)
- Dictionary-based (BOSS, cBOSS)
- Frequency-based (RISE — like TimeSeriesForest but with other features)
- Shapelet-based (Shapelet Transform Classifier)
How is a time series used in clustering?
A time series is a sequence of variable values ordered by time. These data are analyzed using a variety of statistical techniques, such as classification, clustering, and anomaly detection. This paper focuses on clustering.
How is the UCR time series classification resource funded?
Last major update, Summer 2015: Early work on this data resource was funded by an NSF Career Award 0237918, and it continues to be funded through NSF IIS-1161997 II and NSF IIS 1510741. Discontinued! There is now a larger archive here.
How are data points divided into clusters in machine learning?
Clustering is a well-known unsupervised machine learning method for dividing data points (i.e., observations) into groups (called “clusters”) such that observations within the same cluster tend to be more similar (according to a pre-specified criteria) than those in different clusters ( Wu & Kumar, 2009 ).
How big is the UCR time series archive?
Discontinued! There is now a larger archive here. We suggest you begin by reading the briefing document in PDF or PowerPoint, which also contains the password. Then you can download the entire archive (about 350mb in zipped format).