Confusion matrix for verb agreement errors used in the artificial errors method. Based on the training data from the FCE corpus. S and INF stand for third person singular and bare verb form, respectively. The left column shows the correct verb. Each row shows the author's verb choices for that label and Prob(source|label). The numbers next to the targets show the count of the label (or source) in the data set.
Label . | Sources . | |||
---|---|---|---|---|
S . | INF . | WAS . | WERE . | |
(627) . | (2,205) . | (601) . | (149) . | |
S (640) | 0.961 | 0.039 | - | - |
INF (2,192) | 0.005 | 0.995 | - | - |
WAS (593) | - | - | 0.997 | 0.003 |
WERE (157) | - | - | 0.064 | 0.936 |
Label . | Sources . | |||
---|---|---|---|---|
S . | INF . | WAS . | WERE . | |
(627) . | (2,205) . | (601) . | (149) . | |
S (640) | 0.961 | 0.039 | - | - |
INF (2,192) | 0.005 | 0.995 | - | - |
WAS (593) | - | - | 0.997 | 0.003 |
WERE (157) | - | - | 0.064 | 0.936 |