Named Entity Recognition on ADE (test)
96.31F1 ScoreBiLSTM-CNN-Char
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| BiLSTM-CNN-CharEmbeddings=BERT, Evaluation Protocol=relax, Averaging=macro2022.01 | 96.31 | 94.47 | 98.22 | — | |
| BiLSTM-CNN-CharEmbeddings=BERT, Evaluation Protocol=relax, Averaging=micro2022.01 | 96.21 | 94.37 | 98.13 | — | |
| BiLSTM-CNN-CharEmbeddings=GLoVe, Evaluation Protocol=relax, Averaging=macro2022.01 | 94.27 | 93.77 | 94.8 | — | |
| BiLSTM-CNN-CharEmbeddings=GLoVe, Evaluation Protocol=relax, Averaging=micro2022.01 | 94.12 | 93.59 | 94.66 | — | |
| BiLSTM-CNN-CharEmbeddings=BERT, Evaluation Protocol=strict, Averaging=macro2022.01 | 91.75 | 90 | 93.56 | — | |
| BiLSTM-CNN-CharEmbeddings=BERT, Evaluation Protocol=strict, Averaging=micro2022.01 | 91.36 | 89.6 | 93.18 | — | |
| Yan et al. (2021)Averaging=macro, Evaluation Protocol=strict2022.01 | 91.3 | — | — | — | |
| TriMFEvaluation protocol=boundary, Averaging method=macro-average2021.01 | 90.38 | 89.5 | 91.29 | — | |
| Table-Sequence EncodersAveraging=macro-averaged2020.10 | 89.7 | — | — | — | |
| Eberts and UlgesAveraging=macro-averaged2020.10 | 89.3 | — | — | — | |
| SpERTEvaluation protocol=strict, Averaging method=macro-average2021.01 | 89.28 | 88.99 | 89.59 | — | |
| BiLSTM-CNN-CharEmbeddings=GLoVe, Evaluation Protocol=strict, Averaging=macro2022.01 | 88.78 | 88.32 | 89.26 | — | |
| BiLSTM-CNN-CharEmbeddings=GLoVe, Evaluation Protocol=strict, Averaging=micro2022.01 | 88.3 | 87.81 | 88.81 | — | |
| Tran and KavuluruAveraging=macro-averaged2020.10 | 87.1 | — | — | — | |
| Multi-head + ATAveraging method=macro-average2021.01 | 86.73 | — | — | — | |
| Bekoulis et al.Averaging=macro-averaged2020.10 | 86.7 | — | — | — | |
| Bekoulis et al.Averaging=macro-averaged2020.10 | 86.4 | — | — | — | |
| Li et al.Averaging=macro-averaged2020.10 | 84.6 | — | — | — | |
| Li et al.Averaging=macro-averaged2020.10 | 79.5 | — | — | — | |
| Bekoulis et al., 2018avalidation=10-fold cross-validation2021.09 | — | — | — | 86.73 | |
| Bekoulis et al., 2018bvalidation=10-fold cross-validation2021.09 | — | — | — | 86.4 | |
| Eberts and Ulges, 2020validation=10-fold cross-validation2021.09 | — | — | — | 89.25 | |
| Li et al., 2016validation=10-fold cross-validation2021.09 | — | — | — | 79.5 | |
| Li et al., 2017validation=10-fold cross-validation2021.09 | — | — | — | 84.6 | |
| Oursvalidation=10-fold cross-validation, model=CLNER2021.09 | — | — | — | 88.3 | |
| Tran and Kavuluru, 2019validation=10-fold cross-validation2021.09 | — | — | — | 87.11 | |
| Wang and Lu, 2020validation=10-fold cross-validation2021.09 | — | — | — | 89.7 | |
| Zhao et al., 2020validation=10-fold cross-validation2021.09 | — | — | — | 89.4 |