shay681 commited on
Commit
2ae4bec
·
verified ·
1 Parent(s): a5d7199

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +48 -3
README.md CHANGED
@@ -1,3 +1,48 @@
1
- ---
2
- license: unlicense
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - he
4
+ tags:
5
+ - legal, verdicts, metadata, hebrew
6
+ pretty_name: find legal clauses in Legal documents
7
+ size_categories:
8
+ - 100K<n<1M
9
+ task_ids:
10
+ - language-modeling
11
+ - masked-language-modeling
12
+ - document-retrieval
13
+ task_categories:
14
+ - text-generation
15
+ - text-retrieval
16
+ ---
17
+ # Dataset Card for Dataset Name
18
+
19
+ ## Dataset Description
20
+
21
+ ### Dataset Summary
22
+ This dataset represents a 2022 snapshot of the Supreme Court of Israel public verdicts and decisions supported by rich metadata. The 5.31GB dataset represents 751,194 documents.
23
+ Overall, the dataset contains 2.68 Gb of text.
24
+ It can be loaded with the dataset package:
25
+ ```
26
+ import datasets
27
+ data = datasets.load_dataset('LevMuchnik/SupremeCourtOfIsrael')
28
+ ```
29
+
30
+ ### Languages
31
+
32
+ The vast majority of the documents in the database are in Hebrew. A small number of documents are in English.
33
+
34
+ ## Dataset Structure
35
+ The dataset is a json lines file with each line corresponding to a single document and containing document identification, text and metadata.
36
+
37
+
38
+ ### Data Fields
39
+ The file contains the following fields:
40
+ - Id - document id
41
+ - text - text of the document. Multiple newlines and other document formating elements (paragraphs,lists, etc.) are preserved.
42
+ - Legal_Clauses_Found - legal clauses extracted from the text column.
43
+
44
+ ### Data Splits
45
+ The entire dataset is qualified as 'train'.
46
+
47
+ ### Citation Information
48
+ Shay Doner