基于知识图谱的抗炎饮食智能问答系统的构建及应用效果研究OA

[Objective]Develop an intelligent question-answering system based on an anti-inflammatory diet knowledge graph to enhance the accessibility and accuracy of nutritional information for individuals with chronic low-grade inflammatory conditions.[Method]By integrating knowledge data from multiple sources including academic papers,nutritional guidelines,and expert consensus,we manually constructed the model layer and data layer of the knowledge graph.This involved defining entity types and relationships between entities,fol‐lowed by knowledge extraction and fusion.The resulting"entity-relationship-entity"triplet data was input into the Neo4j graph database for storage.Based on the anti-inflammatory knowledge graph,we completed knowledge graph query tasks and designed and developed an intelligent anti-inflammatory diet question-answering function through the collaboration of multidisciplinary team members.[Result]The constructed knowledge graph encompasses seven entity types:foods,inflammatory tendencies,food categories,functional effects,nutrients,disease populations,and dietary patterns.It includes six inter-entity relationships and connects 2,869"entity-relationship-entity"knowledge points.Based on this anti-inflammatory diet knowledge graph,a web-based intelligent question-answering system has been developed.[Conclusion]The Anti-Inflammatory Diet Intelligent Question-Answering System enables individuals with chronic low-grade inflammatory conditions to swiftly access evidence-based,professional,and standardized anti-inflammatory dietary information.It provides a scientific and convenient nutritional management tool for those with chronic low-grade inflammatory diseases,offering new technological support for chronic disease prevention and control.