Table 3.
Result table for AID1189 taken as test set for the models prepared by different sets i.e. Set 1, Set 2 and Set 3
| Classifiers | Data set | Accuracy% | Precision % | Recall% | ROC |
|---|---|---|---|---|---|
|
Naïve Bayes |
Set 1 |
49.08 |
53.3 |
36.80 |
50.30% |
| |
Set 2 |
49.28 |
53.7 |
36.29 |
50.60% |
| |
Set 3 |
49.01 |
51.5 |
49 |
55.5% |
|
Random Forest |
Set 1 |
64.65 |
66.4 |
64.7 |
67.3% |
| |
Set 2 |
61.61 |
66.6 |
56.21 |
64.50% |
| |
Set 3 |
62.89 |
64 |
62.9 |
65.60% |
|
J48 |
Set 1 |
63.16 |
68.6 |
57.10 |
64.60% |
| |
Set 2 |
60.39 |
66 |
53.04 |
62.50% |
| |
Set 3 |
61.27 |
62.1 |
62.3 |
60.8% |
|
SMO |
Set 1 |
50.57 |
55.3 |
38.57 |
55.90% |
| |
Set 2 |
57.14 |
63.2 |
46.95 |
57.90% |
| Set 3 | 56.12 | 57 | 56.1 | 61.2% |
It was found that AID1194 classified better on Set 3 with above 90% accuracy.