Design And Implementation Of Questionnaire Investigation System Based On Micro-service Architecture

With the promotion of information science and technology,Internet technology is changing people's way of life.Network questionnaire investigation is a common survey form in daily production and life.In the beginning,few people participated in the network questionnaire investigation,so the questionnaire investigation system is relatively simple.With the development of the Internet,more and more people participate in the questionnaire investigation,and the traditional single building could not meet people's needs.Therefore,the system is developed by separating the front and back ends in this paper.BootStrap framework is used in the front end,and microservice architecture is used in the back end to develop the system which makes the system highly available,highly concurrent,easy to expand,and so on.A questionnaire investigation system based on the microservice architecture is implemented.In this paper,the requirements of the questionnaire investigation system are analyzed,and the overall architecture of the system is designed by Spring Cloud.In Spring Cloud,Eureka is used to realize the registration and discovery of services,Zuul is used to realize the service gateway,and Feign is used to realize the communication between services.Feign also integrates the Ribbon load balancing function to automatically balance remote calls..Each microservice in the project is built using Spring Boot and developed in Java.Based on the overall framework,the business functions of the system are divided into user management microservice,user registration and login microservice,personal information management microservice,questionnaire design and release microservice,questionnaire query and answer microservice,and questionnaire statistical analysis microservice.In the questionnaire design and release microservice,homonym errors may occur due to negligence when users input the questionnaire content.In this paper,the model combined2-gram and 3-gram is used to check and correct the homonym errors in the questionnaire content,and the error correction results are presented in the front interface.To prove the validity of the combined model for checking and correcting the homonym errors,this paper compares the combined model with 3-gram model and 2-gram model.The contents of Guangming Daily website from 2016 to 2020 and People's Daily Website from 2019 to 2020 are obtained as experimental data.The same training set is used to train the models,and the same test set is used to test the models.The results show that the combined 2-gram and 3-gram model is better than the 3-gram and the 2-gram model.Finally,this paper completes the functions of the system,which can meet the basic needs of the questionnaire investigation.It solves the problems of system access bottleneck and poor function expansion caused by too many visitors in the single structure and also can check and correct the homonym errors in the questionnaire content.