This is a BERT Base model for emotion analysis in Japanese additionally fine-tuned for emotion detection and classification.
The model was based on tohoku-nlp/bert-base-japanese, and later finetuned on a dataset containing 10 labels of emotional blog posts.
The dataset was composed of about 1,000 sentences, with about 100 sentences each for each emotion category.
emotion_mapping = { 0: 'amaze', 1: 'anger', 2: 'dislike', 3: 'excite', 4: 'fear', 5: 'joy', 6: 'like', 7: 'relief', 8: 'sad', 9: 'shame' }
emotion_mapping = { 0: '驚き', 1: '怒り', 2: 'いや', 3: '昂り', 4: '怖がり', 5: '喜び', 6: '好き', 7: '安らぎ', 8: '悲しみ', 9: '恥ずかしい' }