Update README.md
Browse files
README.md
CHANGED
@@ -57,13 +57,13 @@ task_ids:
|
|
57 |
|
58 |
- **Homepage:**
|
59 |
- **Repository:**
|
60 |
-
- **Paper:**
|
61 |
- **Leaderboard:**
|
62 |
-
- **Point of Contact:**
|
63 |
|
64 |
### Dataset Summary
|
65 |
|
66 |
-
[
|
67 |
|
68 |
### Supported Tasks and Leaderboards
|
69 |
|
@@ -71,13 +71,36 @@ task_ids:
|
|
71 |
|
72 |
### Languages
|
73 |
|
74 |
-
|
75 |
|
76 |
## Dataset Structure
|
77 |
|
78 |
### Data Instances
|
79 |
|
80 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
81 |
|
82 |
### Data Fields
|
83 |
|
@@ -97,7 +120,9 @@ task_ids:
|
|
97 |
|
98 |
#### Initial Data Collection and Normalization
|
99 |
|
100 |
-
[
|
|
|
|
|
101 |
|
102 |
#### Who are the source language producers?
|
103 |
|
@@ -111,7 +136,7 @@ task_ids:
|
|
111 |
|
112 |
#### Who are the annotators?
|
113 |
|
114 |
-
|
115 |
|
116 |
### Personal and Sensitive Information
|
117 |
|
|
|
57 |
|
58 |
- **Homepage:**
|
59 |
- **Repository:**
|
60 |
+
- **Paper:** coming soon
|
61 |
- **Leaderboard:**
|
62 |
+
- **Point of Contact:** [Marco Wrzalik](mailto:marco.wrzalik@hs-rm.de)
|
63 |
|
64 |
### Dataset Summary
|
65 |
|
66 |
+
German Legal Sentences (GLS) is an automatically generated training dataset for semantic sentence matching and citation recommendation in the domain in german legal documents. It follows the concept of weak supervision, where imperfect labels are generated using multiple heuristics. For this purpose we use a combination of legal citation matching and BM25 similarity. The contained sentences and their citations are parsed from real judicial decisions provided by [Open Legal Data](http://openlegaldata.io/) (https://arxiv.org/abs/2005.13342).
|
67 |
|
68 |
### Supported Tasks and Leaderboards
|
69 |
|
|
|
71 |
|
72 |
### Languages
|
73 |
|
74 |
+
The text in this dataset is in english
|
75 |
|
76 |
## Dataset Structure
|
77 |
|
78 |
### Data Instances
|
79 |
|
80 |
+
```
|
81 |
+
{'query.doc_id': 28860,
|
82 |
+
'query.ref_ids': [6215, 248, 248],
|
83 |
+
'query.sent_id': 304863,
|
84 |
+
'query.text': 'Zudem ist zu berücksichtigen , dass die Vollverzinsung nach '
|
85 |
+
'[REF] i. V. m. [REF] gleichermaßen zugunsten wie zulasten des '
|
86 |
+
'Steuerpflichtigen wirkt , sodass bei einer Überzahlung durch '
|
87 |
+
'den Steuerpflichtigen der Staat dem Steuerpflichtigen neben '
|
88 |
+
'der Erstattung ebenfalls den entstandenen potentiellen Zins- '
|
89 |
+
'und Liquiditätsnachteil in der pauschalierten Höhe des [REF] '
|
90 |
+
'zu ersetzen hat , unabhängig davon , in welcher Höhe dem '
|
91 |
+
'Berechtigten tatsächlich Zinsen entgangen sind .',
|
92 |
+
'related.doc_id': 56348,
|
93 |
+
'related.ref_ids': [248, 6215, 62375],
|
94 |
+
'related.sent_id': 558646,
|
95 |
+
'related.text': 'Ferner ist zu berücksichtigen , dass der Zinssatz des [REF] '
|
96 |
+
'im Rahmen des [REF] sowohl für Steuernachforderung wie auch '
|
97 |
+
'für Steuererstattungen und damit gleichermaßen zugunsten wie '
|
98 |
+
'zulasten des Steuerpflichtigen wirkt , Vgl. BVerfG , '
|
99 |
+
'Nichtannahmebeschluss vom [DATE] [REF] , juris , mit der '
|
100 |
+
'Folge , dass auch Erstattungsansprüche unabhängig davon , ob '
|
101 |
+
'und in welcher Höhe dem Berechtigten tatsächlich Zinsen '
|
102 |
+
'entgangen sind , mit monatlich 0,0 % verzinst werden .'}
|
103 |
+
```
|
104 |
|
105 |
### Data Fields
|
106 |
|
|
|
120 |
|
121 |
#### Initial Data Collection and Normalization
|
122 |
|
123 |
+
The documents we take from [Open Legal Data](http://openlegaldata.io/) (https://arxiv.org/abs/2005.13342) are first preprocessed by removing line breaks, enumeration characters and headings. Afterwards we parse legal citations using hand-crafted regular expressions. Each citation is split into it components and normalized, thus different variants of the same citation are matched together. For instance, "§211 Absatz 1 des Strafgesetzbuches" is normalized to "§ 211 Abs. 1 StGB". Every time we discover an unknown citation, we assign an unique id to it. We use these ids to replace parsed citations in the document text with a simple reference tag containing this id (e.g `[REF321]`). At the same time we parse dates and replace them with the date tag `[DATE]`. Both remove dots which can may be confused with the end of a sentence, which makes the next stage easier.
|
124 |
+
|
125 |
+
We use [SoMaJo](https://github.com/tsproisl/SoMaJo) to perform sentence tokenizing on the pre-processed documents. Each sentence that does not contain at least one legal citation is discarded. For the rest we assign sentence ids, remove all reference ids from them as well as any contents in braces (braces often contain large enumerations of citations and their sources). At the same time we keep track of the corresponding document from which a sentence originates and which references occur in it.
|
126 |
|
127 |
#### Who are the source language producers?
|
128 |
|
|
|
136 |
|
137 |
#### Who are the annotators?
|
138 |
|
139 |
+
The annotations are machine-generated.
|
140 |
|
141 |
### Personal and Sensitive Information
|
142 |
|