The Design of Recommending Diagnosis in the Management System of Physical Health Examination

Different doctor may give different diagnosis descriptions on the same result in the medical management system, some even not standardized and lots of the diagnosis are same, doctorpsilas repeated input is inefficiency and can not solve the customerpsilas queuing problem. The traditional K-nearest neighbor algorithm has the drawback of slow recommend while the sample space is large. This paper adopt two-level K-nearest neighbor algorithm to reduce time complexity of the recommendation algorithm in order to improve the efficiency of doctorpsilas input and the standardize description of the item, implement the initiative recommend of the medical item diagnosis.