Table 1 Experimental results on synthetic datasets.
From: MultiFacTV: module detection from higher-order time series biological data
Low noise level(0.005) | ||||||||
|---|---|---|---|---|---|---|---|---|
3-module | dataset | 5-module | dataset | 8-module | dataset | 10-module | dataset | |
NMI | Fscore | NMI | Fscore | NMI | Fscore | NMI | Fscore | |
EDISA | 0.5923 | 0.8273 | 0.4997 | 0.7554 | 0.6025 | 0.8100 | 0.4512 | 0.6709 |
MetaFac | 0.9117 | 0.9787 | 0.8473 | 0.9472 | 0.6181 | 0.7717 | 0.5485 | 0.7160 |
MultiFacTV | 0.9874 | 0.9982 | 0.9936 | 0.9986 | 0.8273 | 0.9140 | 0.8035 | 0.8844 |
Middle noise level(0.01) | ||||||||
3-module | dataset | 5-module | dataset | 8-module | dataset | 10-module | dataset | |
NMI | Fscore | NMI | Fscore | NMI | Fscore | NMI | Fscore | |
EDISA | 0.4200 | 0.7136 | 0.2907 | 0.6381 | 0.6670 | 0.8654 | 0.3142 | 0.6027 |
MetaFac | 0.9312 | 0.9830 | 0.4710 | 0.6916 | 0.5444 | 0.7051 | 0.4146 | 0.6037 |
MultiFacTV | 0.9920 | 0.9987 | 0.9898 | 0.9978 | 0.8928 | 0.9552 | 0.7801 | 0.8678 |
High noise level(0.02) | ||||||||
3-module | dataset | 5-module | dataset | 8-module | dataset | 10-module | dataset | |
NMI | Fscore | NMI | Fscore | NMI | Fscore | NMI | Fscore | |
EDISA | 0.4493 | 0.7189 | 0.2514 | 0.5923 | 0.2055 | 0.4355 | 0.1496 | 0.4411 |
MetaFac | 0.9260 | 0.9793 | 0.5318 | 0.6804 | 0.2727 | 0.5222 | 0.2479 | 0.4233 |
MultiFacTV | 0.9914 | 0.9985 | 0.9656 | 0.9898 | 0.7757 | 0.8723 | 0.6565 | 0.7747 |