基于医学社交媒体数据的多模态知识图谱构建

A multi-modal knowledge graph construction method is proposed based on medical social media data. By using semantic association, word frequency statistics, sentiment analysis, visual entity recognition and other technologies, this paper analyzes and summarized multi-modal knowledge from medical social media data, extracts three categories of modal information including domain knowledge, text knowledge and visual knowledge, and then establishes semantic association between the existing ontology and these three categories of modal knowledge to improve the integrity and intelligence of the knowledge graph. Multi-modal knowledge fusion lays a foundation to effectively improve the intelligence of medical expert system.