Knowledge Graph Based Chicken Disease Diagnosis Question Answering System

Shushu Gu, Jing Wang, Shaoqiu Zheng, Yanling Pan, Guoxin Jiang, Wenwen Dai, Ziqi Cheng, Delong Chen

Abstract: With the rapid development of natural language processing technology and knowledge graph technology, knowledge graph based intelligent question answering systems are being increasingly applied in various fields and industries. In the poultry industry, it is of great importance for farmers to promptly obtain scientific information of poultry disease diagnosis and curing measurements, where knowledge graph based question answering systems can contribute richly. In this paper, we design and implement Knowledge Graph based Chicken Disease Diagnosis Question Answering System (CDD-QAS) via deep learning models. We build a knowledge graph for chicken disease diagnosis, which contains 28 common chicken diseases and their corresponding symptoms, prevention and curing measures. In the construction of intelligent question answering system, we use BERT-TextCNN to realize the task of intention recognition and use BiLSTM-CRF to realize the task of entity recognition. Experimental results show that our proposed system can achieve better performance than other models, and possess great interactivity and accuracy. The proposed system can make great contribution to poultry industry and sets a good example of applying knowledge graph and deep learning methods in building question answering system.