Evaluation of 100 randomly sampled variation nuclei for training splits of the WSJ, ATB, and FTB. Corpus positions indicates the number of corpus positions in the sample (a variation nucleus by definition appears in at least two corpus positions). Nuclei per tree is the average nuclei per syntactic tree in the corpus, a statistic that gives a rough estimate of variability across the corpus. The type-level error rate indicates the number of variation nuclei for which at least one error existed. The token-level error rate indicates the ratio of errors to corpus positions. We computed 95% confidence intervals for the type-level error rate.
. | Corpus Positions . | Nuclei Per Tree . | Error % . | Type 95% . | |
---|---|---|---|---|---|
Type . | Token . | Confidence Interval . | |||
PTB (2-21) | 750 | 0.565 | 16.0% | 4.10% | [8.80%, 23.2%] |
ATB (train) | 658 | 0.830 | 26.0% | 4.00% | [17.4%, 34.6%] |
FTB (train) | 668 | 1.10 | 28.0% | 9.13% | [19.2%, 36.8%] |
. | Corpus Positions . | Nuclei Per Tree . | Error % . | Type 95% . | |
---|---|---|---|---|---|
Type . | Token . | Confidence Interval . | |||
PTB (2-21) | 750 | 0.565 | 16.0% | 4.10% | [8.80%, 23.2%] |
ATB (train) | 658 | 0.830 | 26.0% | 4.00% | [17.4%, 34.6%] |
FTB (train) | 668 | 1.10 | 28.0% | 9.13% | [19.2%, 36.8%] |