Table 3 Semantic frame induction results achieved using different community detection algorithms
From: Semantic frame induction through the detection of communities of verbs and their arguments
d | Edges | CC | Clusters | Purity F1 | BCubed F1 | |
|---|---|---|---|---|---|---|
Chinese Whispers | 0.47 | 37,913 | 0.94 | 22.97 ±0.41 | 95.83 ±0.28 | 93.66 ±0.32 |
Louvain Method | 0.43 | 35,491 | 0.92 | 23.00 ±0.00 | 95.83 ±0.00 | 93.66 ±0.00 |
Label Propagation | 0.41 | 33,711 | 0.90 | 23.00 ±0.00 | 95.83 ±0.00 | 93.66 ±0.00 |
Greedy Modularity | 0.31 | 21,753 | 0.84 | 33.00 ±0.00 | 87.15 ±0.00 | 80.45 ±0.00 |