Research Article
Social Network Community-Discovery Algorithm Based on a Balance Factor
Table 2
Experimental results of the community-discovery algorithms on small graphs (F1 and NMI).
| Dataset | Criterion | FG | GN | k-means | Equation (20) | SLC | MIGA | BComd |
| Karate | F1 | 0.971 | 0.970 | 0.879 | 1.0 | 0.971 | 1.0 | 1.0 | NMI | 0.837 | 0.836 | 0.666 | 1.0 | 0.837 | 1.0 | 1.0 |
| Dolphin | F1 | 0.937 | 0.980 | 0.770 | 0.961 | 0.980 | 0.965 | 1.0 | NMI | 0.652 | 0.890 | 0.417 | 0.814 | 0.890 | 0.814 | 1.0 |
| Football | F1 | 0.528 | 0.802 | 0.730 | 0.859 | 0.846 | 0.864 | 0.842 | NMI | 0.697 | 0.878 | 0.822 | 0.865 | 0.793 | 0.916 | 0.847 |
| Political books | F1 | 0.725 | 0.808 | 0.655 | 0.829 | 0.798 | 0.797 | 0.782 | NMI | 0.568 | 0.568 | 0.454 | 0.597 | 0.584 | 0.585 | 0.637 |
|
|
The bold values show best performance.
|