Research And Implementation Of Diagnosis And Treatment Interactive System Based On Pattern Matching Algorithm

Nowadays,with the rise of the Internet,the spread of undesirable information on the Internet and the increasing demand for medical care,the functions and performance of many medical interaction systems cannot meet the needs of users,so this thesis designs and develops a medical interaction system based on the Spring Boot framework,the Cloud Medical Record System,to address the shortcomings of current medical interaction systems.In addition,the current medical interaction system in the market rarely takes into account the possibility of sensitive information transmission in the medical interaction system,so this paper further adds a sensitive semantic detection module based on the original functions of the cloud medical record system for the existence of sensitive information and gives a specific design plan.The main work of the paper has the following aspects.(1)In view of the shortcomings of the traditional diagnosis and treatment interactive system,in order to reduce the configuration work of developers,this paper uses the springboot framework to replace the traditional J2 EE development framework.Based on the idea of separation of front and back ends,the system is designed and developed based on MVC mode.In order to facilitate the subsequent system maintenance and upgrading,Vue framework was used for client development.At the same time,the element UI was used to design the front-end interface,and then the performance of the system was improved to a certain extent through using redis,mysql,mybatis and other technologies,which provided a solid foundation for the subsequent response to the situation of large amount of concurrency,information sharing,instant messaging and other functional development.(2)After constructing a sensitive lexicon for the cloud medical record system,a DAT-ACSVM method for sensitive semantic detection is proposed.Firstly,the tree building process of Double-Array Trie(DAT)and the double-array construction method are optimized and its optimization effect is tested,and the AC(Aho-Corasick)algorithm is introduced to improve it for its shortcomings in the case of multi-mode matching,and the performance of the DATAC algorithm is tested.After that,a Support Vector Machine(SVM)-based method for detecting sentiment tendency of sensitive text is designed considering the problem of sentiment tendency of sensitive text,and it is compared with the method based on the plain Bayesian model to highlight the advantages of the SVM model.Finally,the DAT-AC algorithm is combined with the SVM-based sensitive text sentiment tendency detection method to support the implementation of the sensitive word detection module function of the cloud medical record system.(3)According to the design,the functional modules of the cloud medical record system were implemented and functional and non-functional tests were conducted on its functional modules,with emphasis on the test results of the sensitive semantic module.First,the cloud medical record system was built by writing codes from the login module,personal information management module,medical record management module,administrator review module,message module and sensitive semantic detection module respectively.Then,functional tests are conducted for their modules,and the detection process of the sensitive semantic module applying the DAT-AC-SVM method is described in particular.The results show that the proposed DAT-AC-SVM method can effectively fulfill the functional requirements of the module.Finally,the non-functional aspects of the cloud medical record system are also tested.