Automatic Construction of Knowledge Graph of Tea Diseases and Pests
Qiang Huang, Youzhi Tao, Shitao Ding, Yongbo Liu, Francesco Marinello · Annals of Computer Science and Information Systems · 2023
Tea production involves several stages, usually pests and diseases can negatively impact the quality of tea and reduce the harvest, limiting the industry's development.Therefore, it is important to unify the knowledge on tea pests and diseases.Unfortunately, the current knowledge graph construction for tea pests and diseases relies mainly on semi-automated and manual methods, resulting in inefficiency and failing to meet production demands.This research combines three model of Bidirectional Encoder Representation from Transformers, Bi-directional Long Short-Term Memory, Conditional Random Fields for joint extraction of data.The model abbreviated as BERT-BiLSTM-CRF, using the model can automatically generates the triplets, and then store them in the Neo4j database.The study shows that this model has improved accuracy compared to previous methods, and provides effective support for scientific management and production services of tea pests and diseases.The research offers a reference for quickly constructing knowledge graphs in the crop domain.