"Early Detection of Anthrax from Survey-Based Symptom Data"

"This dataset presents field-collected livestock disease cases acquired by veterinary students through structured questionnaires administered to farmers. The data capture farmer-observed clinical signs - such as fever, convulsions, salivation, recumbency, and trembling, etc., which were subsequently organized into structured feature representations and annotated with disease labels, including anthrax. Given the rarity and field-level difficulty of identifying confirmed anthrax cases, the dataset comprises 10 high-quality samples. Despite its size, it reflects authentic reporting conditions in rural settings, encompassing variability and subjectivity in symptom description. Besides, it is cross-checked by experts in the veterinary field. The dataset is intended to support methodological research in few-shot learning, robust classification under data scarcity, and the development of practical decision-support tools for veterinary epidemiology. Lastly, these samples can also be used for testing an already built system."