Segmentation of Continuous Data Streams Based on a Change Detection Methodology - Advanced Techniques in Knowledge Discovery and Data Mining - page 116

Class = 1: 1
Region = 0: 1
Region = 1: 1
Region = 2: 1
Region = 3: 1
Population = 0: 1
Population = 1: 0
Population = 2: 1
Population = 3: 1
Population = 4: 1
Region = 4: 1
Population = 0: 1
Divisions = 0: 1
Divisions = 1: 2
Population = 3: 2
Divisions = 1: 1
Divisions = 2: 2
Population = 4: 1
Region = 5: 1

Class = 0: 1
Region = 0: 1
Region = 1: 1
Population = 0: 2
Divisions = 0: 2
Divisions = 1: 0
Population = 1: 1
Divisions = 0: 1
Divisions = 1: 1
Divisions = 2: 2
Population = 2: 1
Divisions = 1: 2
Divisions = 2: 1
Divisions = 3: 1

Region = 2: 2
Population = 0: 2
Population = 1: 1
Population = 2: 2
Divisions = 2: 2
Divisions = 3: 1

Region = 3: 1
Population = 0: 1
Population = 1: 1
Population = 2: 2
Population = 3: 1
Population = 4: 2
Region = 4: 1
Population = 0: 2
Population = 1: 1
Population = 2: 1
Divisions = 1: 1
Divisions = 2: 1
Divisions = 3: 1
Population = 3: 0
Population = 4: 1
Population = 5: 1
Region = 5: 2
Population = 0: 2
Divisions = 0: 2
Divisions = 1: 1
Population = 1: 1

Fig. 4.2. Decision tree based on years 1996 - 2000 from “Dropout” database.