A knowledge graph construction method and system for personalized health consultation
By generating user feature vectors and dynamically updating knowledge subgraphs, the problems of low reasoning efficiency and inflexible data management in personalized health consultation are solved, and efficient, safe, and interpretable personalized health consultation services are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGSU MAYTECH MEDICAL TECH CO LTD
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies in personalized health consultation suffer from problems such as low efficiency of global knowledge graph reasoning, inflexible management of dynamic health data, and lack of reasoning basis for consultation answers.
By generating user feature vectors, filtering user-specific knowledge subgraphs, combining dynamic health data to generate personalized fusion knowledge graphs, and performing multi-hop reasoning on the personalized fusion knowledge graphs to generate personalized consultation answers, while dynamically updating user-specific knowledge subgraphs.
It improves reasoning efficiency, enables flexible integration of static knowledge and real-time data, protects user privacy and security, enhances the interpretability and credibility of services, and ensures the accuracy of personalized services.
Smart Images

Figure CN122334423A_ABST