Recurrence Based Similarity Identification of Climate Data
Table 3
Dataset description.
Dataset
Groups
Number of samples in each group
Remark
Description
Synthetic datasets
C5Y120
5
120 points
5 (parameters)
Generated using tentSim fNonlinear package
C10Y120
10
120 points
10 (parameters)
Generated using logisticSim fNonlinear package
C15Y120
15
120 points
15 (parameters)
Generated using henonSim fNonlinear package
Real datasets
US Climate data
9
120 (years)
Samples contain regions’ monthly average data
Climate data having monthly average temperature data
China Climate data
8
50 (years)
Samples contain regions’ monthly average data
Climate data having monthly average temperature data
Heart rate
4
10 (people)
Samples are grouped based on age
There are a total of forty 120 min ECG recordings for different age groups
Voice data
6
10 (people)
Samples are grouped based on healthy status and diseases
The duration of these vowel samples was 3 s for healthy voices and 1 s for pathological voices. A 25 or 50 kHz sampling rate was employed to pathological and healthy voices, respectively. All the files were downsampled to 25 kHz and a single 400 ms frame of analysis was considered for each signal