Datasets:
Upload folder using huggingface_hub
Browse files- .DS_Store +0 -0
- .gitattributes +23 -0
- ML-dataset/corpus.json +3 -0
- ML-dataset/test-queries.json +3 -0
- ML-dataset/train-queries.json +3 -0
- README.md +118 -3
- filtered/corpus_filtered_AUT.json +3 -0
- filtered/corpus_filtered_ESP.json +3 -0
- filtered/corpus_filtered_FRA.json +3 -0
- filtered/corpus_filtered_IRL.json +3 -0
- filtered/corpus_filtered_ITA.json +3 -0
- filtered/directives_filtered_DE.json +3 -0
- filtered/directives_filtered_EN.json +3 -0
- filtered/directives_filtered_ES.json +3 -0
- filtered/directives_filtered_FR.json +3 -0
- filtered/directives_filtered_IT.json +3 -0
- raw/corpus_AUT.json +3 -0
- raw/corpus_ESP.json +3 -0
- raw/corpus_FRA.json +3 -0
- raw/corpus_IRL.json +3 -0
- raw/corpus_ITA.json +3 -0
- raw/directives_DE.json +3 -0
- raw/directives_EN.json +3 -0
- raw/directives_ES.json +3 -0
- raw/directives_FR.json +3 -0
- raw/directives_IT.json +3 -0
- raw/transpositions_ESP.json +0 -0
- raw/transpositions_FRA.json +0 -0
- raw/transpositions_IRL.json +0 -0
- raw/transpositions_ITA.json +0 -0
- raw/transpotitions_AUT.json +0 -0
.DS_Store
ADDED
|
Binary file (14.3 kB). View file
|
|
|
.gitattributes
CHANGED
|
@@ -58,3 +58,26 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 58 |
# Video files - compressed
|
| 59 |
*.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 60 |
*.webm filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
# Video files - compressed
|
| 59 |
*.mp4 filter=lfs diff=lfs merge=lfs -text
|
| 60 |
*.webm filter=lfs diff=lfs merge=lfs -text
|
| 61 |
+
ML-dataset/corpus.json filter=lfs diff=lfs merge=lfs -text
|
| 62 |
+
ML-dataset/test-queries.json filter=lfs diff=lfs merge=lfs -text
|
| 63 |
+
ML-dataset/train-queries.json filter=lfs diff=lfs merge=lfs -text
|
| 64 |
+
filtered/corpus_filtered_AUT.json filter=lfs diff=lfs merge=lfs -text
|
| 65 |
+
filtered/corpus_filtered_ESP.json filter=lfs diff=lfs merge=lfs -text
|
| 66 |
+
filtered/corpus_filtered_FRA.json filter=lfs diff=lfs merge=lfs -text
|
| 67 |
+
filtered/corpus_filtered_IRL.json filter=lfs diff=lfs merge=lfs -text
|
| 68 |
+
filtered/corpus_filtered_ITA.json filter=lfs diff=lfs merge=lfs -text
|
| 69 |
+
filtered/directives_filtered_DE.json filter=lfs diff=lfs merge=lfs -text
|
| 70 |
+
filtered/directives_filtered_EN.json filter=lfs diff=lfs merge=lfs -text
|
| 71 |
+
filtered/directives_filtered_ES.json filter=lfs diff=lfs merge=lfs -text
|
| 72 |
+
filtered/directives_filtered_FR.json filter=lfs diff=lfs merge=lfs -text
|
| 73 |
+
filtered/directives_filtered_IT.json filter=lfs diff=lfs merge=lfs -text
|
| 74 |
+
raw/corpus_AUT.json filter=lfs diff=lfs merge=lfs -text
|
| 75 |
+
raw/corpus_ESP.json filter=lfs diff=lfs merge=lfs -text
|
| 76 |
+
raw/corpus_FRA.json filter=lfs diff=lfs merge=lfs -text
|
| 77 |
+
raw/corpus_IRL.json filter=lfs diff=lfs merge=lfs -text
|
| 78 |
+
raw/corpus_ITA.json filter=lfs diff=lfs merge=lfs -text
|
| 79 |
+
raw/directives_DE.json filter=lfs diff=lfs merge=lfs -text
|
| 80 |
+
raw/directives_EN.json filter=lfs diff=lfs merge=lfs -text
|
| 81 |
+
raw/directives_ES.json filter=lfs diff=lfs merge=lfs -text
|
| 82 |
+
raw/directives_FR.json filter=lfs diff=lfs merge=lfs -text
|
| 83 |
+
raw/directives_IT.json filter=lfs diff=lfs merge=lfs -text
|
ML-dataset/corpus.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d646eb61cde50198e3a93f19e6eddd0bc155b16e9eb4045aa973f7763fa1286d
|
| 3 |
+
size 3045510396
|
ML-dataset/test-queries.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2d4d8734d8f5425f4f0efd711996a53bbd3e48ccc81422413bde81abb4295dd
|
| 3 |
+
size 21010721
|
ML-dataset/train-queries.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:922aa04207a933fa210fa5e4fe26c0747d5ef1423faec5f5d15acff619031c24
|
| 3 |
+
size 176387715
|
README.md
CHANGED
|
@@ -1,3 +1,118 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
--
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
annotations_creators:
|
| 3 |
+
- no-annotation
|
| 4 |
+
language_creators:
|
| 5 |
+
- found
|
| 6 |
+
language:
|
| 7 |
+
- it
|
| 8 |
+
- fr
|
| 9 |
+
- es
|
| 10 |
+
- en
|
| 11 |
+
- de
|
| 12 |
+
license: cc-by-4.0
|
| 13 |
+
multilinguality:
|
| 14 |
+
- multilingual
|
| 15 |
+
size_categories:
|
| 16 |
+
- 100K<n<1M
|
| 17 |
+
source_datasets:
|
| 18 |
+
- original
|
| 19 |
+
task_categories:
|
| 20 |
+
- text-retrieval
|
| 21 |
+
- sentence-similarity
|
| 22 |
+
task_ids:
|
| 23 |
+
- semantic-search
|
| 24 |
+
- document-retrieval
|
| 25 |
+
pretty_name: European National Implementing Measures Dataset (ENIMD)
|
| 26 |
+
tags:
|
| 27 |
+
- legal
|
| 28 |
+
- european-union
|
| 29 |
+
- harmonization
|
| 30 |
+
- legislation
|
| 31 |
+
---
|
| 32 |
+
|
| 33 |
+
# ENIMD: European National Implementing Measures Dataset
|
| 34 |
+
|
| 35 |
+
**ENIMD** is a large-scale multilingual dataset designed for **legal semantic search** and **harmonization analysis**. It pairs **European Directives (EUDs)** with their corresponding **National Implementing Measures (NIMs)** across five Member States.
|
| 36 |
+
|
| 37 |
+
The dataset enables the training of models to automatically identify national laws that implement EU directives, distinguishing them from domestic legislation that does not.
|
| 38 |
+
|
| 39 |
+
## 📄 Paper
|
| 40 |
+
**Pairing EU directives and their national implementing measures: A dataset for semantic search**
|
| 41 |
+
*Roger Ferrod, Denys Amore Bondarenko, Davide Audrito, Giovanni Siragusa*
|
| 42 |
+
Published in **Computer Law & Security Review**, Volume 51, 2023.
|
| 43 |
+
|
| 44 |
+
[**Read the Paper**](https://doi.org/10.1016/j.clsr.2023.105862) | [**GitHub Repository**](https://github.com/rogerferrod/ENIMD)
|
| 45 |
+
|
| 46 |
+
## 💾 Dataset Structure
|
| 47 |
+
|
| 48 |
+
The dataset is organized into three components, catering to different research needs:
|
| 49 |
+
|
| 50 |
+
### 1. `ML-dataset`
|
| 51 |
+
A shuffled, machine-learning-ready collection of articles split into **Train** and **Test** sets.
|
| 52 |
+
* **Content:** Pairs of EUD articles (Queries) and National Law articles (Documents).
|
| 53 |
+
* **Labels:** Includes `positive` examples (NIMs) and `negative` examples (irrelevant national laws).
|
| 54 |
+
* **Structure:** Articles are labeled with the CELEX number, country code and transposition hash.
|
| 55 |
+
* **Preprocessing:** Filtered using an IDF-based method to remove boilerplate text (e.g., entry into force dates, financial clauses).
|
| 56 |
+
|
| 57 |
+
### 2. `filtered`
|
| 58 |
+
The parsed collection of articles where irrelevant/boilerplate provisions have been removed using the method described in the paper. Useful for analysis without the noise of administrative clauses.
|
| 59 |
+
|
| 60 |
+
### 3. `raw`
|
| 61 |
+
The full parsed collection of Directives and National Laws in their original structure (articles/paragraphs), without any filtering.
|
| 62 |
+
|
| 63 |
+
## 📊 Statistics
|
| 64 |
+
|
| 65 |
+
The dataset covers legislation from five EU Member States:
|
| 66 |
+
|
| 67 |
+
| Country | Language | EUD Articles (Queries) | National Corpus Articles |
|
| 68 |
+
|:----------| :--- |:-----------------------|:-------------------------|
|
| 69 |
+
| Italy | Italian | 11,514 | 135,221 |
|
| 70 |
+
| France | French | 11,386 | 236,762 |
|
| 71 |
+
| Spain | Spanish | 11,249 | 209,795 |
|
| 72 |
+
| Ireland | English | 11,344 | 157,601 |
|
| 73 |
+
| Austria | German | 11,837 | 199,781 |
|
| 74 |
+
| **Total** | **Multilingual** | **57,330** | **939,160** |
|
| 75 |
+
|
| 76 |
+
* **Total Directives:** 906
|
| 77 |
+
* **Total National Documents:** 9,016
|
| 78 |
+
* **Ratio:** ~88% of the national corpus consists of "irrelevant" laws (negative examples), providing a realistic retrieval challenge.
|
| 79 |
+
|
| 80 |
+
## 💻 Usage
|
| 81 |
+
|
| 82 |
+
You can load the dataset using the Hugging Face `datasets` library.
|
| 83 |
+
|
| 84 |
+
```python
|
| 85 |
+
from datasets import load_dataset
|
| 86 |
+
|
| 87 |
+
# Load the ML-ready dataset
|
| 88 |
+
dataset = load_dataset("rogerferrod/ENIMD", data_dir="ML-dataset")
|
| 89 |
+
|
| 90 |
+
# Example: Inspect the first training example
|
| 91 |
+
print(dataset['train'][0])
|
| 92 |
+
```
|
| 93 |
+
|
| 94 |
+
## ⚖️ Legal Harmonization Task
|
| 95 |
+
|
| 96 |
+
The primary task is Semantic Search / Retrieval:
|
| 97 |
+
|
| 98 |
+
1. Query: An article from an EU Directive.
|
| 99 |
+
2. Target: The specific article(s) in National Law that implement that directive.
|
| 100 |
+
3. Challenge: The model must retrieve the correct implementation from a pool of ~900k national articles, most of which are unrelated.
|
| 101 |
+
|
| 102 |
+
## 📚 Citation
|
| 103 |
+
|
| 104 |
+
If you use this dataset in your research, please cite the original paper:
|
| 105 |
+
|
| 106 |
+
```bibtex
|
| 107 |
+
@article{FERROD2023105862,
|
| 108 |
+
title = {Pairing EU directives and their national implementing measures: A dataset for semantic search},
|
| 109 |
+
journal = {Computer Law & Security Review},
|
| 110 |
+
volume = {51},
|
| 111 |
+
pages = {105862},
|
| 112 |
+
year = {2023},
|
| 113 |
+
issn = {2212-473X},
|
| 114 |
+
doi = {https://doi.org/10.1016/j.clsr.2023.105862},
|
| 115 |
+
url = {https://www.sciencedirect.com/science/article/pii/S0267364923000729},
|
| 116 |
+
author = {Roger Ferrod and Denys Amore Bondarenko and Davide Audrito and Giovanni Siragusa}
|
| 117 |
+
}
|
| 118 |
+
```
|
filtered/corpus_filtered_AUT.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cf1acd8bb8af7dcc88998a3631ca5bfe4e93594d41263778b2591f66639b0fe
|
| 3 |
+
size 565582920
|
filtered/corpus_filtered_ESP.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68e7258bd8fda8beac952c080690e3053d6a3f34c84b963ad27f2f9410c1af68
|
| 3 |
+
size 349276536
|
filtered/corpus_filtered_FRA.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:533a6ae3095ab4bde2ab7b4c6ba987bbf1206f08e3500eda079beb78fc61a11e
|
| 3 |
+
size 341819001
|
filtered/corpus_filtered_IRL.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cd1b6451534e44a8edf6f48610d343d1f20afb3a534d115127fa9d7dbdc93d9
|
| 3 |
+
size 172776213
|
filtered/corpus_filtered_ITA.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf32d2d1e45223e7cff6c4bd13647591c005d7a42ef30807096f27e08a2cc101
|
| 3 |
+
size 236612619
|
filtered/directives_filtered_DE.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:703c233dc30613d933181b19ad3c60d85bfe0bc00836d7c82f58f4928941d695
|
| 3 |
+
size 24706048
|
filtered/directives_filtered_EN.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:291fd0d559e86844145840acc56902fa8d6335199ed94c959aff8f051cf93041
|
| 3 |
+
size 22315679
|
filtered/directives_filtered_ES.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb7a258c9f6a5149142ec4728362b6ef0258ce16c8a328e3b66eeffa5f9c85a9
|
| 3 |
+
size 24718461
|
filtered/directives_filtered_FR.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9cbb24b0951c56f4cf38e1d2f10607e0742922f19d0d0361e95dcc393e3b9b7c
|
| 3 |
+
size 24727053
|
filtered/directives_filtered_IT.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a17d762e860b44d5c50c73a8a0abd2df311e95709bc7acd0fc881ba7f2255248
|
| 3 |
+
size 23909677
|
raw/corpus_AUT.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:145d8ab977791abbb9082b8b5dee13fcfa79ac854d21cbdb0b75cd26a8cc095f
|
| 3 |
+
size 620659836
|
raw/corpus_ESP.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60790fa1ce398a759d1f9e3ab3f75da9d2f9e79851fd77801d67023e38d17a04
|
| 3 |
+
size 479939579
|
raw/corpus_FRA.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:043cedb3bccc01a4c20f79b300cc7341790cf15019f1496fc1a4c234fd7a13b6
|
| 3 |
+
size 366511332
|
raw/corpus_IRL.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d46b954c3f1b39e6ebe6cd21f81176ed668ca81430b5a519cf36b6507ee15555
|
| 3 |
+
size 206505618
|
raw/corpus_ITA.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2a3ac1a09ea2f6a9397e47031d999ce43f631298eda4ab295a7f3b6680471ea
|
| 3 |
+
size 272390525
|
raw/directives_DE.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:687656b9a9bc0a19ec30d85dee985dd7fadd6ae46c5e3c8fa1b28a67bcc62a95
|
| 3 |
+
size 25911925
|
raw/directives_EN.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:065bc7663feb7fd715446fd0af2b35f2f763d0cf0c65ff3c8ef2b3526362370a
|
| 3 |
+
size 23775109
|
raw/directives_ES.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d46a20cd27009cae3d2d7774cb24873e09d621373470eaa4814d7bedfed3dc4b
|
| 3 |
+
size 26245478
|
raw/directives_FR.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d27f86744f780589d87fd5bfb9cce70fe41c719b73fb54e669f83d960388ed3
|
| 3 |
+
size 26126238
|
raw/directives_IT.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0f5dab8559e9a74b38bef36f3bbb1f223e9d83004b88169d471402f4c1ea240
|
| 3 |
+
size 25266830
|
raw/transpositions_ESP.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
raw/transpositions_FRA.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
raw/transpositions_IRL.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
raw/transpositions_ITA.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
raw/transpotitions_AUT.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|