Time Series Segmentation for Optimal Multi-Segment Linear Regression (OMSLR) Paper
- simup - Simulation Perfect
- 10 series data
- 80 data points with 5 trends
- data range: [0, 100]
- simun - Simulation with Noise
- based on simup, add +/- 5 noise.
- simul - Simulation Long Series
- 1 series data
- 1000 data points with 20 trends
- ucr - UC Riverside dataset
- Opendata from UCR time series dataset
- 12 samples from different type
- source link
- Description of samples as follow
| Exp# | Type | Name | Length |
|---|---|---|---|
| 0 | ECG | ECGFiveDays | 136 |
| 1 | EOG | EOGVerticalSignal | 1250 |
| 2 | EPG | InsectEPGRegularTrain | 601 |
| 3 | Hemodynamics | PigCVP | 2000 |
| 4 | HRM | Fungi | 201 |
| 5 | Image | MiddlePhalanxOutlineAgeGroup | 80 |
| 6 | Motion | UWaveGestureLibraryAll | 945 |
| 7 | Power | PowerCons | 144 |
| 8 | Sensor | Plane | 144 |
| 9 | Simulated | CBF | 128 |
| 10 | Spectro | Ham | 431 |
| 11 | Spectrum | Rock | 2844 |