结合信息增益率的异构医疗数据库字段标识系统设计

为解决现阶段医疗机构中不同科室数据异构,导致标识准确率低的问题,提出并设计了结合信息增益率的异构医疗数据库字段标识系统。给出设计系统整体结构,分析设计系统主要模块功能。对异构医疗信息进行数据标识化处理,令原本没有语义信息的异构医疗数据变成可处理数据。在此基础上,将信息增益率看作节点分割的优度衡量指标建立多特征函数,依据特征函数标识异构医疗数据库信息时,通过标识规则对数据种类进行判断,获取全部医疗信息数据库所有数据字段标识结果,完成异构医疗数据库字段标识系统设计。实验结果表明,所设计系统标识结果准确,操作性能优,标识时延低。

Design of Heterogeneous Medical Database Field Identification System Combined with Information Gain Rate

In order to solve the problem of heterogeneous data of different departments in medical institutions at present, which leads to low accuracy of identification, a field identification system of heterogeneous medical database combining information gain rate is proposed and designed. The whole structure of the design system is given, and the main module functions of the design system are analyzed. The heterogeneous medical information is processed by data identification, which makes the heterogeneous medical data without semantic information processable. On the basis of this, the information gain rate is regarded as an index to measure the degree of superiority of node segmentation. When the heterogeneous medical database information is identified according to the feature function, the classification of the data is judged by the identification rules, and the whole medical treatment is obtained. All the data fields of the information database are identified, and the field identification system of the heterogeneous medical database is designed. The experimental results show that the designed system has the advantages of accurate identification, good operation performance and low marking delay.