Table 2 Frequency of occurrence of the labels on the data splits of the Vent dataset after pre-processing. The proportion of the total number of instances within the sample is in parenthesis
From: LEIA: Linguistic Embeddings for the Identification of Affect
| Â | Train | Development | User Test | Temporal test | Random test |
|---|---|---|---|---|---|
Sadness | 1,712,985 (27%) | 199,890 (28%) | 262,999 (27%) | 293,993 (30%) | 264,906 (27%) |
Anger | 1,517,282 (24%) | 147,778(21%) | 224,997 (23%) | 205,598 (21%) | 226,068 (23%) |
Fear | 1,341,624 (21%) | 138,929 (20%) | 198,264 (21%) | 185,461 (19%) | 201,563 (21%) |
Affection | 979,019 (15%) | 144,175 (20%) | 161,018 (17%) | 191,022 (20%) | 158,017 (16%) |
Happiness | 795,363 (13%) | 74,369 (11%) | 118,290 (12%) | 91,127 (9%) | 116,647 (12%) |
Total | 6,346,273 | 705,141 | 965,568 | 967,201 | 967,201 |