Roya Javadi, Angelica Lim
The portrayal of negative emotions such as anger can vary widely between cultures and contexts, depending on the acceptability of expressing full-blown emotions rather than suppression to maintain harmony. The majority of emotional datasets collect data under the broad label ``anger", but social signals can range from annoyed, contemptuous, angry, furious, hateful, and more. In this work, we curated the first in-the-wild multicultural video dataset of emotions, and deeply explored anger-related emotional expressions by asking culture-fluent annotators to label the videos with 6 labels and 13 emojis in a multi-label framework. We provide a baseline multi-label classifier on our dataset, and show how emojis can be effectively used as a language-agnostic tool for annotation.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Text Classification | MFA | F-F1 score (Comb.) | 0.34 | MLKNN |
| Text Classification | MFA | F-F1 score (NA) | 0.42 | MLKNN |
| Text Classification | MFA | F-F1 score (Persian) | 0.4 | MLKNN |
| Text Classification | MFA | V-F1 score (Comb.) | 0.39 | MLKNN |
| Text Classification | MFA | V-F1 score (NA) | 0.42 | MLKNN |
| Text Classification | MFA | V-F1 score (Persian) | 0.4 | MLKNN |
| Text Classification | MFA | F-F1 score (Comb.) | 0.33 | CC - XGB |
| Text Classification | MFA | F-F1 score (NA) | 0.42 | CC - XGB |
| Text Classification | MFA | F-F1 score (Persian) | 0.28 | CC - XGB |
| Text Classification | MFA | V-F1 score (Comb.) | 0.36 | CC - XGB |
| Text Classification | MFA | V-F1 score (NA) | 0.4 | CC - XGB |
| Text Classification | MFA | V-F1 score (Persian) | 0.33 | CC - XGB |
| Emotion Classification | MFA | F-F1 score (Comb.) | 0.34 | MLKNN |
| Emotion Classification | MFA | F-F1 score (NA) | 0.42 | MLKNN |
| Emotion Classification | MFA | F-F1 score (Persian) | 0.4 | MLKNN |
| Emotion Classification | MFA | V-F1 score (Comb.) | 0.39 | MLKNN |
| Emotion Classification | MFA | V-F1 score (NA) | 0.42 | MLKNN |
| Emotion Classification | MFA | V-F1 score (Persian) | 0.4 | MLKNN |
| Emotion Classification | MFA | F-F1 score (Comb.) | 0.33 | CC - XGB |
| Emotion Classification | MFA | F-F1 score (NA) | 0.42 | CC - XGB |
| Emotion Classification | MFA | F-F1 score (Persian) | 0.28 | CC - XGB |
| Emotion Classification | MFA | V-F1 score (Comb.) | 0.36 | CC - XGB |
| Emotion Classification | MFA | V-F1 score (NA) | 0.4 | CC - XGB |
| Emotion Classification | MFA | V-F1 score (Persian) | 0.33 | CC - XGB |
| Classification | MFA | F-F1 score (Comb.) | 0.34 | MLKNN |
| Classification | MFA | F-F1 score (NA) | 0.42 | MLKNN |
| Classification | MFA | F-F1 score (Persian) | 0.4 | MLKNN |
| Classification | MFA | V-F1 score (Comb.) | 0.39 | MLKNN |
| Classification | MFA | V-F1 score (NA) | 0.42 | MLKNN |
| Classification | MFA | V-F1 score (Persian) | 0.4 | MLKNN |
| Classification | MFA | F-F1 score (Comb.) | 0.33 | CC - XGB |
| Classification | MFA | F-F1 score (NA) | 0.42 | CC - XGB |
| Classification | MFA | F-F1 score (Persian) | 0.28 | CC - XGB |
| Classification | MFA | V-F1 score (Comb.) | 0.36 | CC - XGB |
| Classification | MFA | V-F1 score (NA) | 0.4 | CC - XGB |
| Classification | MFA | V-F1 score (Persian) | 0.33 | CC - XGB |