File size: 4,681 Bytes
8a77321
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
---
configs:
- config_name: document_metadata
  data_files:
  - split: train
    path: data/document_metadata/document_metadata.parquet
- config_name: spans
  data_files:
  - split: train
    path: data/spans/spans.parquet
- config_name: relations
  data_files:
  - split: train
    path: data/relations/relations.parquet
task_categories:
- token-classification
language:
- fr
tags:
- inception
- uima
- annotation
---
 
# Dataset Card for pseudo
 
## Dataset Description
 
- **Homepage:** - **Point of Contact:**
 
### Dataset Summary
 
ce truc decrit un truc
 
pseudo is a subpart of the PARHAF corpus, an open French corpus of human-authored clinical reports of fictional patients.
 
It was created to support the development and evaluation of clinical NLP systems for {use_case_short_description}.
 
Each patient record was:
 
- written by a senior medical resident
- reviewed by another senior medical resident, from the same specialty
- annotated by a specialist of the use case
- curated by another specialist of the use case
 
#### Data statistics
 
### Indicateurs globaux

| Indicateur | Valeur |
| --- | --- |
| Nombre de fichiers JSON | 846 |
| Total annotations | 11610 |
| Longueur moyenne des documents | 4009 caractères |
| Seuil 60% (annotations) | 6966 |

### Répartition par type du dataset complet

| Type | Annotations | % du total |
| --- | --- | --- |
| EntiteAnonymisation | 11610 | 100.00% |

### Répartition par champ renseigné du dataset complet

| Type.Champ | Occurrences | % du total annot. |
| --- | --- | --- |
| EntiteAnonymisation.Categorie | 11610 | 100.00% |
| EntiteAnonymisation.CodeType | 2 | 0.02% |
| EntiteAnonymisation.RoleLOC | 148 | 1.27% |
| EntiteAnonymisation.RoleNUM | 19 | 0.16% |
| EntiteAnonymisation.RolePER | 4145 | 35.70% |

### Résultats de la division Train / Test

| Indicateur | TRAIN | TOTAL |
| --- | --- | --- |
| Nombre de fichiers | 509 | 509 |
| Nombre d'annotations | 6976 | 6976 |
| Pourcentage du dataset | 60.09% | 100.00% |
| Longueur moy. documents (chars) | 4021 |  |

### Répartition par type Train

| Type | TRAIN | % train du type |
| --- | --- | --- |
| EntiteAnonymisation | 6976 | 60.09% |

### Répartition par champ renseigné Train

| Type.Champ | TRAIN | % train du champ |
| --- | --- | --- |
| EntiteAnonymisation.Categorie | 6976 | 60.09% |
| EntiteAnonymisation.CodeType | 0 | 0.00% |
| EntiteAnonymisation.RoleLOC | 87 | 58.78% |
| EntiteAnonymisation.RoleNUM | 8 | 42.11% |
| EntiteAnonymisation.RolePER | 2466 | 59.49% |

### Répartition par valeur Train

| Type.Champ.Valeur | TRAIN | % train |
| --- | --- | --- |
| EntiteAnonymisation.Categorie.DateIdentifiante | 2880 | 60.67% |
| EntiteAnonymisation.Categorie.Nom | 1398 | 59.19% |
| EntiteAnonymisation.RolePER.Patient | 1235 | 58.59% |
| EntiteAnonymisation.RolePER.PersonnelSante | 1205 | 60.49% |
| EntiteAnonymisation.Categorie.Prenom | 1068 | 59.90% |
| EntiteAnonymisation.Categorie.SituationFamiliale | 970 | 59.51% |
| EntiteAnonymisation.Categorie.IdentiteSocialePatient | 523 | 61.82% |
| EntiteAnonymisation.Categorie.Ville | 54 | 58.70% |
| EntiteAnonymisation.RoleLOC.Patient | 46 | 66.67% |
| EntiteAnonymisation.RolePER.Autre | 26 | 57.78% |
| EntiteAnonymisation.RoleLOC.Hopital | 22 | 50.00% |
| EntiteAnonymisation.Categorie.Pays | 28 | 65.12% |
| EntiteAnonymisation.RoleLOC.Autre | 19 | 54.29% |
| EntiteAnonymisation.Categorie.DateNaissancePatient | 17 | 51.52% |
| EntiteAnonymisation.Categorie.DateNonIdentifiante | 19 | 59.38% |
| EntiteAnonymisation.Categorie.NumeroTelephone | 8 | 42.11% |
| EntiteAnonymisation.RoleNUM.PersonnelSante | 5 | 50.00% |
| EntiteAnonymisation.Categorie.Adresse | 3 | 30.00% |
| EntiteAnonymisation.Categorie.NationalitePatient | 6 | 85.71% |
| EntiteAnonymisation.RoleNUM.Hopital | 2 | 33.33% |
| EntiteAnonymisation.Categorie.Url | 2 | 50.00% |
| EntiteAnonymisation.RoleNUM.Autre | 1 | 33.33% |
| EntiteAnonymisation.Categorie.CodeIdentifiant | 0 | 0.00% |
| EntiteAnonymisation.CodeType.NIPouIPP | 0 | 0.00% |
| EntiteAnonymisation.CodeType.INS | 0 | 0.00% |

 
#### Data Origin
 
The clinical reports are extracted from
 
## Structure des fichiers
 
| Config | Contenu |
|---|---|
| `document_metadata` | Titre, texte brut, annotation document-level |
| `spans` | Annotations Span avec leurs attributs |
| `relations` | Relations entre spans |
 
## Chargement
 
```python
from datasets import load_dataset
 
metadata  = load_dataset("abdellaouic/test", name="document_metadata", split="train")
spans     = load_dataset("abdellaouic/test", name="spans",             split="train")
relations = load_dataset("abdellaouic/test", name="relations",         split="train")
```