Update README.md
Browse files
README.md
CHANGED
@@ -90,10 +90,10 @@ EUROPA is a dataset designed for training and evaluating multilingual keyphrase
|
|
90 |
|
91 |
## Dataset Structure
|
92 |
|
93 |
-
- **celex_id:**
|
94 |
- **lang:** ISO 639-1 language code
|
95 |
-
- **input:** judgment transcription or
|
96 |
-
- **keyphrases:** keyphrases drafted by the CJEU
|
97 |
|
98 |
## Citation
|
99 |
|
|
|
90 |
|
91 |
## Dataset Structure
|
92 |
|
93 |
+
- **celex_id:** CELEX identifier inherited from CJEU. Different translated versions of the same judgment share the same celex_id.
|
94 |
- **lang:** ISO 639-1 language code
|
95 |
+
- **input:** judgment transcription or translation
|
96 |
+
- **keyphrases:** reference keyphrases drafted by the CJEU. Phrases are separated by semicolons.
|
97 |
|
98 |
## Citation
|
99 |
|