基于知识图谱的食品领域数智化研究进展

Research Progress on Digitalization and Intelligence in Food Domain Based on Knowledge Graphs

随着大数据和云计算等技术的发展,食品领域的数据规模正在以惊人的速度增长。这些数据不仅来源多样、结构复杂,且缺乏统一的术语标准,给食品数据的有效整合和利用带来了挑战。知识图谱作为实现通用人工智能的重要基石,以其在数据整合和语义理解方面的能力,为食品领域的数据组织、管理与应用提供了支持。通过总结近年来知识图谱在食品领域应用的研究成果,介绍了食品领域知识图谱的构建方法,涵盖本体构建、知识抽取、知识融合和加工等关键环节;重点梳理了当前食品领域知识图谱在食品营养与健康、食品创新与研发、食品安全与溯源3个方面的数智化应用;基于食品领域知识图谱的发展现状,从多模态数据融合技术、大语言模型构建以及食品领域工业设备智能化发展角度展望了食品领域知识图谱的未来发展方向。

With the development of technologies such as big data and cloud computing,the scale of data in the food domain is growing at an astonishing rate.These data not only come from diverse sources and have complex structures,but also lack standardized terminology,which poses challenges to the effective integration and utilization of food-related data.Knowledge graphs,as fundamental cornerstone of achieving general artificial intelligence,provides support for the organization and management of food data and its higher-level applications in terms of integration and semantic understanding.By summarizing recent research achievements of knowledge graphs in the food domain,the construction methods of knowledge graphs in food domain was reviewed,covering key steps such as ontology construction,knowledge extraction,knowledge fusion,and processing.The current applications of knowledge graphs in the food domain,particularly in three areas,food nutrition and health,food innovation and research,and food safety and traceability.Based on current state of development of knowledge graphs in the food domain,incorporating multimodal data fusion technology,large language model construction,and the intelligentization of industrial equipment in the food field,the future development directions of knowledge graphs in food domain were anticipated.