Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -24,9 +24,6 @@ Irony classification task using soft labels (i.e., distribution of annotations)
|
|
24 |
MultiPICo is a multilingual corpus containing texts in different varieties of each language (see column "language_variety")
|
25 |
|
26 |
## Data Instances
|
27 |
-
Total amount of instances: 94,342
|
28 |
-
|
29 |
-
Total number of annotators: 506
|
30 |
|
31 |
- Spanish: 4,683 instances and 122 annotators
|
32 |
- English: 2,999 instances and 74 annotators
|
@@ -38,12 +35,15 @@ Total number of annotators: 506
|
|
38 |
- Italian: 1,000 instances and 24 annotators
|
39 |
- Hindi: 786 instances and 24 annotators
|
40 |
|
|
|
|
|
|
|
|
|
41 |
## Data Fields
|
42 |
MultiPICo is structured as follows:
|
43 |
|
44 |
-
in rows, the annotation of each annotator (identified with a “annotator_id”)
|
45 |
-
|
46 |
-
in columns, the various information about the target text annotated by the user (post_id, post, reply_id, reply, language, and language_variety), and the metadata about annotators (age, sex, ethnicity, and so on).
|
47 |
|
48 |
## Data Splits
|
49 |
The corpus is not split in training and validation/test sets.
|
|
|
24 |
MultiPICo is a multilingual corpus containing texts in different varieties of each language (see column "language_variety")
|
25 |
|
26 |
## Data Instances
|
|
|
|
|
|
|
27 |
|
28 |
- Spanish: 4,683 instances and 122 annotators
|
29 |
- English: 2,999 instances and 74 annotators
|
|
|
35 |
- Italian: 1,000 instances and 24 annotators
|
36 |
- Hindi: 786 instances and 24 annotators
|
37 |
|
38 |
+
Total amount of instances: 94,342
|
39 |
+
|
40 |
+
Total number of annotators: 506
|
41 |
+
|
42 |
## Data Fields
|
43 |
MultiPICo is structured as follows:
|
44 |
|
45 |
+
- in rows, the annotation of each annotator (identified with a “annotator_id”)
|
46 |
+
- in columns, the various information about the target text annotated by the user (post_id, post, reply_id, reply, language, and language_variety), and the metadata about annotators (age, sex, ethnicity, and so on).
|
|
|
47 |
|
48 |
## Data Splits
|
49 |
The corpus is not split in training and validation/test sets.
|