地球资源数据云——数据资源详情
推文情绪
该数据集《Tweet Emotions》主要用于多分类任务，数据形态以图像为主，应用场景偏向医疗健康。 题目说明：Recognise emotion using tweet 任务类型：图像多分类。 建议流程：先检查类别分布与脏样本，再用迁移学习（如 ResNet/EfficientNet）建立基线。 评估建议：使用分层切分或交叉验证，优先关注 F1、Recall、AUC 等分类指标。 可用文件：test.csv, train.csv, val.csv。 This dataset is designed for Facial Emotion Recognition (FER) tasks and contains labeled face images representing a range of human emotions. Facial emotion recognition is a key problem in computer vision and affective computing, with applications in human–computer interaction, mental health analysis, surveillance systems, and AI - driven user experience enhancement. The dataset is curated to support image classification and deep learning workflows, enabling researchers and practitioners to train, evaluate, and benchmark models for emotion detection from facial expressions. The dataset structure and labeling are inspired by widely used FER benchmarks and is suitable for both academic research and practical machine learning projects Emotion Categories Each entry in this dataset consists of a text segment representing a Twitter message and a corresponding label indicating the predominant emotion conveyed. The emotions are classified into six categories: sadness (0), joy (1), love (2), anger (3), fear (4), and surprise (5).
摘要概览
该数据集《Tweet Emotions》主要用于多分类任务，数据形态以图像为主，应用场景偏向医疗健康。 题目说明：Recognise emotion using tweet
任务类型：图像多分类。
建议流程：先检查类别分布与脏样本，再用迁移学习（如 ResNet/EfficientNet）建立基线。
评估建议：使用分层切分或交叉验证，优先关注 F1、Recall、AUC 等分类指标。
可用文件：test.csv, train.csv, val.csv。
This dataset is designed for Facial Emotion Recognition (FER) tasks and contains labeled face images representing a range of human emotions.
Facial emotion recognition is a key problem in computer vision and affective computing, with applications in human–computer interaction, mental health analysis, surveillance systems, and AI - driven user experience enhancement.
The dataset is curated to support image classification and deep learning workflows, enabling researchers and practitioners to train, evaluate, and benchmark models for emotion detection from facial expressions.
The dataset structure and labeling are inspired by widely used FER benchmarks and is suitable for both academic research and practical machine learning projects
Emotion Categories Each entry in this dataset consists of a text segment representing a Twitter message and a corresponding label indicating the predominant emotion conveyed. The emotions are classified into six categories: sadness (0), joy (1), love (2), anger (3), fear (4), and surprise (5).