Research Article
A New Approach for Construction of Geodemographic Segmentation Model and Prediction Analysis
Table 7
Accuracy and error rates of different classification techniques.
| Technique | Accuracy | Error rate |
| CART decision tree | 75.9% (759/1000) | 24.1% (241/1000) | Random tree | 80.5% (805/1000) | 19.5% (195/1000) | Gradient boost | 81.8% (818/1000) | 18.2% (182/1000) | AdaBoost | 80.6% (806/1000) | 19.4% (194/1000) | Extra tree | 80.1% (801/1000) | 19.9% (199/1000) | SVM | 81.0% (810/1000) | 19.0% (190/1000) | ANN | 80.9% (809/1000) | 19.1% (191/1000) | Naïve Bayes | 77.1% (771/1000) | 22.9% (229/1000) | kNN | 80.9% (809/1000) | 19.1 (191/1000) |
|
|