method: TEKLIA DAN (HTR) + SpaCy (NER)2023-01-13
Authors: Solène Tarride, Mélodie Boillet, Christopher Kermorvant
Affiliation: TEKLIA
Description: We use DAN for HTR on records, and then we apply SpaCy for NER.
Two SpaCy models are trained: one for the person, the other for the category.
method: TEKLIA DAN (HTR+NER)2023-02-07
Authors: Solène Tarride, Mélodie Boillet, Christopher Kermorvant
Affiliation: TEKLIA
Description: We train DAN for HTR and NER on records.
We use a unique tag combining the category and person information (ex: [name_wife]Maria).
method: TEKLIA Kaldi + Flair2021-02-22
Authors: Marie-Laurence Bonhomme, Christopher Kermorvant
Affiliation: Teklia
Description: We used existing libraries (Kaldi for the HTR, and a number of NER tools ; here we submit the results obtained with FLAIR) on the IEHHR 2017 task (basic track) to compare the results in terms of impact of the HTR on the NER results to those observed on other datasets.
| Date | Method | Basic Score | Complete Score | Name | Surname | Location | Occupation | State | Input Type | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2023-01-13 | TEKLIA DAN (HTR) + SpaCy (NER) | 97.13% | 97.11% | 97.88% | 96.97% | 96.56% | 97.43% | 96.99% | REGISTER | |||
| 2023-02-07 | TEKLIA DAN (HTR+NER) | 97.03% | 96.93% | 97.63% | 95.92% | 96.63% | 97.23% | 97.83% | REGISTER | |||
| 2021-02-22 | TEKLIA Kaldi + Flair | 96.96% | 0.00% | 97.70% | 95.18% | 96.26% | 97.53% | 98.22% | LINE | |||
| 2024-02-12 | LITIS DAN (HTR+NER) | 96.80% | 96.84% | 96.30% | 96.60% | 96.29% | 97.34% | 97.57% | PAGE | |||
| 2023-01-13 | TEKLIA PyLaia (HTR) + SpaCy (NER) | 96.58% | 96.58% | 97.36% | 96.34% | 95.82% | 96.16% | 97.12% | REGISTER | |||
| 2023-02-07 | TEKLIA DAN (Key-value HTR+NER) | 96.48% | 96.31% | 97.33% | 95.32% | 95.49% | 97.35% | 97.07% | REGISTER | |||
| 2021-01-21 | InstaDeep-CRNS | 96.25% | 95.54% | 96.58% | 94.60% | 95.81% | 95.92% | 98.19% | REGISTER | |||
| 2023-01-03 | GNN-Transformer | 96.22% | 96.24% | 97.38% | 95.47% | 95.44% | 96.23% | 96.63% | REGISTER | |||
| 2018-06-25 | Naver Labs | 95.46% | 95.03% | 97.01% | 92.73% | 95.03% | 96.43% | 96.41% | LINE | |||
| 2017-07-09 | CITlab ARGUS (with OOV) | 91.94% | 91.58% | 95.14% | 85.78% | 88.43% | 93.08% | 97.54% | LINE | |||
| 2017-07-10 | CITlab ARGUS (with OOV, net2) | 91.63% | 91.19% | 95.09% | 85.84% | 87.32% | 92.96% | 97.19% | LINE | |||
| 2018-10-27 | Joint HTR + NER no postprocessing | 90.59% | 89.40% | 89.94% | 84.07% | 90.71% | 92.10% | 96.59% | LINE | |||
| 2017-07-09 | CITlab ARGUS (without OOV) | 89.54% | 89.17% | 94.37% | 76.54% | 87.65% | 92.66% | 97.43% | LINE | |||
| 2017-07-01 | Baseline HMM | 80.28% | 63.11% | 81.06% | 60.15% | 78.90% | 90.23% | 93.79% | LINE |