Hugging Face's logo

Datasets:
cardiffnlp
/
tweet_topic_single

Tasks:
Text Classification
Modalities:
Text
Sub-tasks:
sentiment-classification
Languages:
English
Size:
10K - 100K
ArXiv:
Libraries:
Datasets
License:
Dataset card Viewer Files Community
1
asahi417 commited on
Commit
4c6566d
·
1 Parent(s): dc1be7c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -13
README.md CHANGED
@@ -69,21 +69,23 @@ print(target_format)
69
 
70
 
71
  ### Data Splits
 
72
  | split | number of texts | description |
73
  |:------------------------|-----:|------:|
74
- | test_2020 | 573 | test dataset from September 2019 to August 2020 |
75
- | test_2021 | 1679 | test dataset from September 2020 to August 2021 |
76
- | train_2020 | 4585 | training dataset from September 2019 to August 2020 |
77
- | train_2021 | 1505 | training dataset from September 2020 to August 2021 |
78
- | train_all | 6090 | combined training dataset of `train_2020` and `train_2021` |
79
- | validation_2020 | 573 | validation dataset from September 2019 to August 2020 |
80
- | validation_2021 | 188 | validation dataset from September 2020 to August 2021 |
81
- | train_random | 4564 | randomly sampled training dataset with the same size as `train_2020` from `train_all` |
82
- | validation_random | 573 | randomly sampled training dataset with the same size as `validation_2020` from `validation_all` |
83
- | test_coling2022_random | 5536 | random split used in the COLING 2022 paper |
84
- | train_coling2022_random | 5731 | random split used in the COLING 2022 paper |
85
- | test_coling2022 | 5536 | temporal split used in the COLING 2022 paper |
86
- | train_coling2022 | 5731 | temporal split used in the COLING 2022 paper |
 
87
  For the temporal-shift setting, model should be trained on `train_2020` with `validation_2020` and evaluate on `test_2021`.
88
  In general, model would be trained on `train_all`, the most representative training set with `validation_2021` and evaluate on `test_2021`.
89
  **IMPORTANT NOTE:** To get a result that is comparable with the results of the COLING 2022 Tweet Topic paper, please use `train_coling2022` and `test_coling2022` for temporal-shift, and `train_coling2022_random` and `test_coling2022_random` fir random split (the coling2022 split does not have validation set).
 
69
 
70
 
71
  ### Data Splits
72
+
73
  | split | number of texts | description |
74
  |:------------------------|-----:|------:|
75
+ | test_2020 | 376 | test dataset from September 2019 to August 2020 |
76
+ | test_2021 | 1693 | test dataset from September 2020 to August 2021 |
77
+ | train_2020 | 2858 | training dataset from September 2019 to August 2020 |
78
+ | train_2021 | 1516 | training dataset from September 2020 to August 2021 |
79
+ | train_all | 4374 | combined training dataset of `train_2020` and `train_2021` |
80
+ | validation_2020 | 352 | validation dataset from September 2019 to August 2020 |
81
+ | validation_2021 | 189 | validation dataset from September 2020 to August 2021 |
82
+ | train_random | 2830 | randomly sampled training dataset with the same size as `train_2020` from `train_all` |
83
+ | validation_random | 354 | randomly sampled training dataset with the same size as `validation_2020` from `validation_all` |
84
+ | test_coling2022_random | 3399 | random split used in the COLING 2022 paper |
85
+ | train_coling2022_random | 3598 | random split used in the COLING 2022 paper |
86
+ | test_coling2022 | 3399 | temporal split used in the COLING 2022 paper |
87
+ | train_coling2022 | 3598 | temporal split used in the COLING 2022 paper |
88
+
89
  For the temporal-shift setting, model should be trained on `train_2020` with `validation_2020` and evaluate on `test_2021`.
90
  In general, model would be trained on `train_all`, the most representative training set with `validation_2021` and evaluate on `test_2021`.
91
  **IMPORTANT NOTE:** To get a result that is comparable with the results of the COLING 2022 Tweet Topic paper, please use `train_coling2022` and `test_coling2022` for temporal-shift, and `train_coling2022_random` and `test_coling2022_random` fir random split (the coling2022 split does not have validation set).