摘要：

Abstract:

【Purpose/significance】Under the premise of insufficient existing information recommendation methods, in-depth analysis

of the advantages of knowledge graphs in health medical community information recommendation and different scenarios in health in⁃

formation recommendation, aiming to provide scenario-based information recommendation for different types of users.【Method/pro⁃

cess】According to the division of the scene, a co-topic recommendation model based on semantics, a personalized recommendation

model based on disease portraits and user interests, and a situational information recommendation model are designed, and specific

models and algorithms are designed and implemented.【Result/conclusion】Construct recommendation models in different scenarios,

reason and output various medical knowledge and information, which can assist patients who desire more accurate online health infor⁃

mation.【Innovation/limitation】Proposed an online medical community information recommendation fusion of knowledge maps and ill⁃

ness profiles, and constructed an online medical community information recommendation program that fused knowledge maps and ill⁃

ness profiles. It is still necessary to further strengthen the application of the integration of knowledge graphs and user portraits.