Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Languages:
French
Size:
10K - 100K
License:
bourdoiscatie
commited on
Commit
•
681fbe8
1
Parent(s):
fc7a1c9
Update README.md
Browse files
README.md
CHANGED
@@ -83,11 +83,28 @@ dataset = load_dataset("CATIE-AQ/universal_dependencies_fr_sequoia_fr_prompt_pos
|
|
83 |
|
84 |
# Citation
|
85 |
## Original data
|
86 |
-
>
|
87 |
-
|
88 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
89 |
|
90 |
## This Dataset
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
91 |
|
92 |
|
93 |
|
|
|
83 |
|
84 |
# Citation
|
85 |
## Original data
|
86 |
+
> @inproceedings{candito:hal-00698938,
|
87 |
+
TITLE = {{Le corpus Sequoia : annotation syntaxique et exploitation pour l'adaptation d'analyseur par pont lexical}},
|
88 |
+
AUTHOR = {Candito, Marie and Seddah, Djam{\'e}},
|
89 |
+
URL = {https://inria.hal.science/hal-00698938},
|
90 |
+
BOOKTITLE = {{TALN 2012 - 19e conf{\'e}rence sur le Traitement Automatique des Langues Naturelles}},
|
91 |
+
ADDRESS = {Grenoble, France},
|
92 |
+
YEAR = {2012},
|
93 |
+
MONTH = Jun,
|
94 |
+
PDF = {https://inria.hal.science/hal-00698938/file/canditoseddah-taln2012-final.pdf},
|
95 |
+
HAL_ID = {hal-00698938},
|
96 |
+
HAL_VERSION = {v1},
|
97 |
+
}
|
98 |
|
99 |
## This Dataset
|
100 |
+
> @misc {centre_aquitain_des_technologies_de_l'information_et_electroniques_2023,
|
101 |
+
author = { {Centre Aquitain des Technologies de l'Information et Electroniques} },
|
102 |
+
title = { DFP (Revision 1d24c09) },
|
103 |
+
year = 2023,
|
104 |
+
url = { https://huggingface.co/datasets/CATIE-AQ/DFP },
|
105 |
+
doi = { 10.57967/hf/1200 },
|
106 |
+
publisher = { Hugging Face }
|
107 |
+
}
|
108 |
|
109 |
|
110 |
|