Table 4.
| Model-Based | k-Means | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
| 1 | 0 | 36 | 1094 | 28 | 0 | 429 | 1 | 0 | 0 |
| 2 | 21 | 95 | 0 | 63 | 17 | 214 | 44 | 860 | 0 |
| 3 | 26 | 101 | 21 | 61 | 117 | 8 | 240 | 18 | 45 |
| 4 | 0 | 87 | 0 | 675 | 52 | 0 | 9 | 12 | 0 |
| 5 | 0 | 18 | 39 | 15 | 52 | 0 | 938 | 9 | 0 |
| 6 | 0 | 727 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 7 | 0 | 0 | 0 | 754 | 20 | 0 | 0 | 0 | 0 |
| 8 | 0 | 12 | 0 | 16 | 888 | 0 | 0 | 23 | 0 |
| 9 | 945 | 22 | 0 | 0 | 0 | 0 | 0 | 4 | 1586 |
Comparison of k-Means to the Model-Based Clustering. While there are some areas where hundreds of members are modeled into the same class, few member counts dominate both the Model-Based and k-Means based classification at the same time, implying that there is some, but not too much, agreement between the two methods.