Design And Implementation Of Teacher Sharing Service Platform

In the process of teaching staff development and construction,local colleges and universities in remote areas are faced with many difficulties in talent introduction.They are faced with difficult problems such as lack of teacher resources,few teachers on assignment,and large turnover of staff.It is urgent to coordinate and dispatch high-quality teachers from other colleges and universities to alleviate the problems and ensure teaching quality.However,the information barrier between colleges and universities has not yet been broken down,which leads to the difficult problem of faculty call.Therefore,it is necessary to reasonably design the teacher-sharing service platform as an information bridge between colleges and universities to help solve the shortage of teachers in local colleges and universities in remote areas.At the same time,the teacher sharing service platform can also expand to assist enterprises,government departments and other demand-side users to solve the problem of teacher resource scheduling.This thesis designs and implements a sharing service platform for teachers to realize the scientific,reasonable and efficient utilization of faculty resources in colleges and universities.The resource-sharing mode is integrated into the functional services of the platform.According to the needs of sharing faculty in colleges and universities,the terminal of the platform is designed as follows: There are three client types: teacher demand response end,school-enterprise demand publishing end and background data management end,which endow users with different platform permissions,provide different platform services for different user groups,and display different client interfaces.In terms of platform functional architecture,B/S architecture is adopted,and the design mode of front and back end separation and decoupling is applied to develop business functions based on the idea of engineering decoupling.According to the functional requirements of the platform,the server uses Java language and builds functional services based on Spring Boot framework.According to the client characteristics of user groups,the front-end of the platform uses Type Script language to build the user interface based on Vue framework.In the request layer,Nginx is used as the request processing server.In the aspect of data storage,My SQL database is used to store persistent data and Redis is used to store cache data.Online chat module based on Socket protocol and multi-thread technology to achieve real-time message transmission among multiple users,using the way of client storage to protect user privacy security.The personalized recommendation module is built based on deep learning neural network.At the recall layer of the module,Spark platform is used to calculate the similarity of the Embedding vector of teacher user information and demand information,so as to complete the rapid recall of platform information.In the sorting layer,DIN deep interest network model is used to fine sort the recalled platform information,and the trained sorting model is deployed through Tensorflow model service,and the obtained model recommendation result is sent back to the service end of the business module through HTTP request.The teacher sharing service platform highly integrates the university teacher resources,improves the utilization rate of university teacher resources,and provides a new solution for the scientific coordination and reasonable distribution of university teacher resources.This is a new attempt to the development of college education brought by the mode of Internet resource sharing.The platform has completed all the development and testing work and deployed on the online server,and will soon carry out user internal testing and platform trial operation.