Research Article
A Modified Decision Tree Algorithm Based on Genetic Algorithm for Mobile User Classification Problem
Table 3
Rules on mobile users generated by the algorithm.
| ID | Rule | Training set accuracy | Test set accuracy |
| 1 |
IF (Wireless cost ≥ 27.7) and (Context = public context) THEN class = Total service | 84.66% | 83.25% | 2 | IF (Wireless cost ≤ 27.7) and (Free part > 13.5) and (Context = public context) THEN class = Plus service | 72.68% | 70.13% | 3 | IF (Wireless cost ≤ 27.7) and (Free part ≤ 13.5) and (Context = private) and (month ≤ 16) THEN class = Basic service | 73.07% | 71.64% | 4 | IF (Wireless cost ≤ 27.7) and (Free part ≤ 13.5) and (Context = private) and (month > 16) THEN class = E-service | 64.87% | 63.79% |
|
|