sentiment analysis

The dataset consists of a multilingual corpus with more than 30,000 tweets in English and Spanish, annotated in five categories: Not Hope, Generalized Hope, Realistic Hope, Unrealistic Hope, and Sarcasm.

The manually annotated emotion recognition dataset, curated in collaboration with local communities, consist of multi-labeled instances drawn from diverse sources, including speeches, social media, news, literature, and reviews. Each instance is labeled by fluent speakers and annotated with six emotion classes: joy, sadness, anger, fear, surprise, disgust, and neutral.

The dataset consists of a set of tweets labeled based on whether they contain hopeful language or not, and if they do, the type of hopeful language they contain.

The dataset consists of audio segments collected from various Spanish-language YouTube channels, labeled with the emotion they convey according to five of Ekman's six basic emotions: anger, disgust, fear, joy, and sadness, as well as a neutral emotion.