Table 5 Statistics on the impact of removing true interactions from the Yeast-D2 Dataset.
Noise | \(Del_{Random}\) | \(Del_{HighDegree}\) | \(Del_{LowDegree}\) | ||||||
|---|---|---|---|---|---|---|---|---|---|
m | \(\vert n \vert _{d=1}\) | \(d_{Avg}\) | m | \(\vert n \vert _{d=1}\) | \(d_{Avg}\) | m | \(\vert n \vert _{d=1}\) | \(d_{Avg}\) | |
0% | 6993 | 92 | 9.6923 | 6993 | 92 | 9.6923 | 6993 | 92 | 9.6923 |
10% | 6332 | 127 | 8.7762 | 6414 | 92 | 8.8898 | 6873 | 95 | 9.5260 |
20% | 5749 | 173 | 7.9681 | 5836 | 92 | 8.0887 | 6752 | 101 | 9.3583 |
30% | 5235 | 178 | 7.2557 | 5257 | 92 | 7.2862 | 6632 | 107 | 9.1920 |
40% | 4721 | 236 | 6.5433 | 4678 | 93 | 6.4837 | 6512 | 119 | 9.0256 |
50% | 4308 | 241 | 5.9709 | 4100 | 100 | 5.6826 | 6391 | 131 | 8.8579 |
