Skip to Main Content
Table 13 

Macro-average recall, precision, and F1 on the development set and test set using the parameters that maximize F1 of the learned edges over the development set.


Development set
Test set
Algorithm
Recall
Precision
F1
Recall
Precision
F1
Linearall 48.1 31.9 36.3 51.7 37.7 40.3 
Linearold 40.3 33.3 34.8 47.2 42.2 41.1 
Gaussianall 41.8 32.4 35.1 48.0 41.1 40.7 
Gaussianold 41.1 31.2 33.9 50.3 39.7 40.5 
Squareall 39.9 32.0 34.1 43.7 39.8 38.9 
Squareold 38.0 31.6 32.9 50.2 41.0 41.3 
Logisticall 34.4 27.6 29.1 39.8 41.7 37.8 
Logisticold 39.3 31.2 33.5 45.4 40.9 39.9 
Bayesall 20.8 33.2 24.5 27.4 46.0 31.7 
Bayesold 20.3 34.9 24.6 26.4 45.4 30.9 

Development set
Test set
Algorithm
Recall
Precision
F1
Recall
Precision
F1
Linearall 48.1 31.9 36.3 51.7 37.7 40.3 
Linearold 40.3 33.3 34.8 47.2 42.2 41.1 
Gaussianall 41.8 32.4 35.1 48.0 41.1 40.7 
Gaussianold 41.1 31.2 33.9 50.3 39.7 40.5 
Squareall 39.9 32.0 34.1 43.7 39.8 38.9 
Squareold 38.0 31.6 32.9 50.2 41.0 41.3 
Logisticall 34.4 27.6 29.1 39.8 41.7 37.8 
Logisticold 39.3 31.2 33.5 45.4 40.9 39.9 
Bayesall 20.8 33.2 24.5 27.4 46.0 31.7 
Bayesold 20.3 34.9 24.6 26.4 45.4 30.9 
Close Modal

or Create an Account

Close Modal
Close Modal