Datasets:
Add files from Clarin Repo
Browse files- ParlaSent_BCS.jsonl +0 -0
- ParlaSent_BCS_test.jsonl +0 -0
- ParlaSent_CZ.jsonl +0 -0
- ParlaSent_EN.jsonl +0 -0
- ParlaSent_EN_test.jsonl +0 -0
- ParlaSent_SK.jsonl +0 -0
- ParlaSent_SL.jsonl +0 -0
- README.txt +37 -0
ParlaSent_BCS.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_BCS_test.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_CZ.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_EN.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_EN_test.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_SK.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
ParlaSent_SL.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
README.txt
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
The multilingual sentiment dataset of parliamentary debates ParlaSent 1.0
|
2 |
+
http://hdl.handle.net/11356/1868
|
3 |
+
|
4 |
+
The dataset consists of five training datasets and two test sets. The test sets have a _test.jsonl suffix.
|
5 |
+
|
6 |
+
The attributes in training data are the following:
|
7 |
+
- sentence - the sentence labeled for sentiment
|
8 |
+
- country - the country of the parliament the sentence comes form
|
9 |
+
- annotator1 - first annotator's annotation
|
10 |
+
- annotator2 - second annotator's annotation
|
11 |
+
- reconciliation - the final label agreed upon after reconciliation
|
12 |
+
- label - three level (positive, negative, neutral) label based on the reconciliation label
|
13 |
+
- document_id - internal identifier of the document the sentence comes form
|
14 |
+
- sentence_id - internal identifier of the sentence inside the document
|
15 |
+
- term - the term of the parliament the sentence comes from
|
16 |
+
- date - the date the sentence was uttered as part of a speech in the parliament
|
17 |
+
- name - name of the MP giving the speech
|
18 |
+
- party - the party of the MP
|
19 |
+
- gender - binary gender of the MP
|
20 |
+
- birth year - year of birth of the MP
|
21 |
+
- split - whether the sentence is to be used as a training, development or testing instance in case evaluation is done of the training portion of the dataset
|
22 |
+
- ruling - whether the MP was in a coalition or an opposition at the time of giving the speech
|
23 |
+
|
24 |
+
The attributes in the test data (_test.jsonl files) are the following:
|
25 |
+
- sentence - the sentence labeled for sentiment
|
26 |
+
- country - the country of the parliament the sentence comes form
|
27 |
+
- annotator1 - first (only) annotator's annotation, used as a final annotation
|
28 |
+
- label - three level (positive, negative, neutral) label based on the annotator1 label
|
29 |
+
- document_id - internal identifier of the document the sentence comes form
|
30 |
+
- sentence_id - internal identifier of the sentence inside the document
|
31 |
+
- term - the term of the parliament the sentence comes from
|
32 |
+
- date - the date the sentence was uttered as part of a speech in the parliament
|
33 |
+
- name - name of the MP giving the speech
|
34 |
+
- party - the party of the MP
|
35 |
+
- gender - binary gender of the MP
|
36 |
+
- birth year - year of birth of the MP
|
37 |
+
- ruling - whether the MP was in a coalition or an opposition at the time of giving the speech
|