A Joint Knowledge Extraction Model for Tobacco Pest and Disease Prevention Based on BERT+BA+CASREL

This is a knowledge extraction dataset for the field of tobacco pests and diseases in China, organized in JSON format. Each data sample contains a descriptive text and its corresponding SPO (subject predicate object) triplet, used to annotate entities and their relationships in the text. The triplet explicitly labels the subject, predicate, object, and their corresponding entity and relationship types.