KoseiUemura commited on
Commit
19add47
·
verified ·
1 Parent(s): d4a4d7e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +90 -1
README.md CHANGED
@@ -1056,4 +1056,93 @@ configs:
1056
  ---
1057
  # Dataset Card for "EmotionAnalysisFinal"
1058
 
1059
- [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1056
  ---
1057
  # Dataset Card for "EmotionAnalysisFinal"
1058
 
1059
+ ---
1060
+ # ---------- REQUIRED METADATA ----------
1061
+ license: cc-by-sa-4.0 # Change if your licence differs
1062
+ pretty_name: EmotionAnalysisFinal
1063
+ tags:
1064
+ - semeval-2025-task-11
1065
+ - cross-lingual-emotion-detection
1066
+ task_categories:
1067
+ - text-classification
1068
+ task_ids:
1069
+ - emotion-detection
1070
+ multilinguality:
1071
+ - multilingual
1072
+ # List every language code that has a config
1073
+ language:
1074
+ - afr # Afrikaans
1075
+ - arq # Algerian Arabic
1076
+ - amh # Amharic
1077
+ - chn # Mandarin Chinese
1078
+ - vmw # Emakhuwa
1079
+ - eng # English
1080
+ - deu # German
1081
+ - hau # Hausa
1082
+ - hin # Hindi
1083
+ - ibo # Igbo
1084
+ - ind # Indonesian
1085
+ - xho # isiXhosa
1086
+ - zul # isiZulu
1087
+ - jav # Javanese
1088
+ - kin # Kinyarwanda
1089
+ - mar # Marathi
1090
+ - ary # Moroccan Arabic
1091
+ - pcm # Nigerian-Pidgin
1092
+ - orm # Oromo
1093
+ - ptbr # Portuguese (Brazil)
1094
+ - ptmz # Portuguese (Mozambique)
1095
+ - ron # Romanian
1096
+ - rus # Russian
1097
+ - som # Somali
1098
+ - esp # Spanish (LatAm)
1099
+ - sun # Sundanese
1100
+ - swa # Swahili
1101
+ - swe # Swedish
1102
+ - tat # Tatar
1103
+ - tir # Tigrinya
1104
+ - ukr # Ukrainian
1105
+ - yor # Yoruba
1106
+
1107
+ # ---------- DATA FILE LOCATIONS ----------
1108
+ # Only dev (=validation) and test splits are available for Track C.
1109
+ configs:
1110
+ {% for subset in [
1111
+ "afr","arq","amh","chn","vmw","eng","deu","hau","hin",
1112
+ "ibo","ind","xho","zul","jav","kin","mar","ary","pcm",
1113
+ "orm","ptbr","ptmz","ron","rus","som","esp","sun","swa",
1114
+ "swe","tat","tir","ukr","yor"
1115
+ ] %}
1116
+ - config_name: {{ subset }}
1117
+ data_files:
1118
+ - split: validation # dev split in the original release
1119
+ path: data/{{ subset }}/dev.csv
1120
+ - split: test
1121
+ path: data/{{ subset }}/test.csv
1122
+ {% endfor %}
1123
+ ---
1124
+
1125
+ # Dataset Card for **EmotionAnalysisFinal**
1126
+
1127
+ `EmotionAnalysisFinal` is the official release of the **SemEval-2025 Task 11, Track C** cross-lingual emotion-detection benchmark.
1128
+ Each language-specific configuration contains:
1129
+
1130
+ | Split | # instances | Notes |
1131
+ |--------------|------------|-------|
1132
+ | `validation` | *≈1 000* | Former “dev” split used during the task |
1133
+ | `test` | *≈1 000* | Gold labels released after evaluation |
1134
+
1135
+ The annotation schema is **multi-label** over six basic emotions: *anger, disgust, fear, joy, sadness,* and *surprise*.
1136
+
1137
+ ## Quick start
1138
+
1139
+ ```python
1140
+ from datasets import load_dataset
1141
+
1142
+ # Load Yoruba test split
1143
+ ds_test = load_dataset(
1144
+ "llama-lang-adapt/EmotionAnalysisFinal",
1145
+ "yor",
1146
+ split="test"
1147
+ )
1148
+ print(ds_test[0])