Design And Implementation Of Campus Second-hand Goods Transaction System Based On Cloud Platform

In recent years,the rapid popularity of the Internet in universities,coupled with the increasing demand for second-hand goods transaction,universities have set up their own second-hand goods transaction platform.With the continuous improvement and promotion of the platform,the data amount of second-hand goods,system users and transaction information on campus is increasing gradually.In order to meet the performance of elastic expansion,high concurrency and load balancing of the platform,this paper is designed with Openstack cloud platform to better realize the functions of commodity recommendation,commodity publishing and commodity search in campus second-hand goods transaction system.The main work of this paper is as follows:(1)the demand analysis and design of the system.First,use case analysis of the user role of the system;then carry on the overall architecture design,database design and UML design;finally,use the RESTful architecture to design the API,so as to realize the Web Service technology.(2)use the independent research and development framework coding to achieve system functions.The development framework uses hierarchical architecture to separate the front and rear ends of the system and reduce the coupling degree of each component.The technologies such as Python,MySQL and Nginx are all open source,and the framework has strong expansibility.This framework is used to develop the secondhand commodity transaction system on campus,which encodes and implements the functions of commodity publishing,commodity search,commodity transaction,commodity recommendation,switching universities,backstage management of university administrators,and super administrator backstage management.(3)research and application of Item-based Collaborative Filtering algorithm.This paper mainly studies the implementation process of Item-based Collaborative Filtering,calculates the similarity between the goods and the interest of the users,generates the list of recommended goods for the users,and uses the precision and recall to evaluate the recommendation function,and uses the Item-based Collaborative Filtering algorithm to recommend the commodity to improve the turnover rate of second-hand goods on campus.(4)the construction and application of cloud platform.This paper builds five components of Openstack's Keystone,Glance,Nova,Neutron and Horizon,launchs the Ubuntu instance on the cloud platform and deploys the system.The characteristics of cloud platform such as elastic expansion,load balancing and on-demand use have been well applied in the system.The system has completed the design,development and cloud deployment,which is testing in the campus of Guizhou University,the expected function is realized and the current operation situation is well.