Design And Implementation Of Agricultural Cultural Tourism Management Platform Based On Personalized Recommendation

Agricultural cultural tourism industry is a new type of tourism industry based on agriculture and oriented to urban tourists,which is helpful to promote the process of rural revitalization.The penetration of Internet technology in the field of agriculture and tourism has given birth to a new form of agricultural cultural tourism.At present,most of China lacks the Internet platform for the agricultural culture tourism industry,and the relevant agricultural websites are mostly focused on the sales of agricultural products.The agricultural culture tourism resources lack the proper channels to promote to the public,and the public cannot accurately obtain the needed agricultural culture tourism information.Therefore,it is necessary to build an agricultural cultural tourism management platform based on personalized recommendation.Aiming at the industrial structure upgrading of "Internet +" agricultural cultural tourism,this paper gives the research and development of agricultural cultural tourism management platform.The agricultural cultural tourism management platform aims to integrate the agricultural cultural tourism resources in various places to realize online browsing of agricultural cultural tourism information,online popularization of agricultural culture,reservation of farmhouse accommodation,appointment of agricultural activities and agricultural products and other services.In order to realize the accurate docking of tourism resources,the platform applies personalized recommendation technology to the development process,and recommends the information that may be of interest to users based on their preferences and specific needs,so as to achieve a win-win result of improving user experience and promoting agricultural tourism market revenue.The specific research results are as follows:(1)Integrated LDA theme model and collaborative filtering recommendation algorithm for hybrid recommendationThrough the analysis and induction of a large number of literatures,the basic principles,advantages and disadvantages of various collaborative filtering recommendation algorithms are compared,and the existing problems of recommendation algorithms are analyzed.LDA theme model was used to generate user-site preference probability matrix to improve the cold start of the platform.Pearson correlation coefficient and Mini Batch k-means clustering algorithm were used to perform clustering filling on the user scoring matrix to improve the sparseness of the scoring matrix.The ebbinghaus forgetting curve was used to improve the time weight,and the minimum diameter circle method was used to improve the distance weight,so as to increase the accuracy of the recommendation results.A hybrid recommendation algorithm for agricultural tourist attractions was constructed by integrating LDA theme model and collaborative filtering.Finally,through the three measurement indexes of accuracy,recall rate and coverage rate,the algorithm in this paper is compared with the traditional recommendation algorithm and the conclusion is drawn.(2)A management platform for agricultural cultural tourism was establishedThe platform adopts the development process of software engineering,develops based on object-oriented analysis and design methods,and constructs three models of function,behavior and data.Firstly,the requirements of the platform client and the management terminal system are analyzed by object-oriented analysis method,and the functional structure of the client and management terminal system is determined by modules.Based on the layered and modular design idea,the four-layer architecture of data persistence layer,business logic layer,control layer and view layer is constructed to determine the platform architecture.Through conceptual data model design and logical data model design to establish the relationship between the data,through the data table structure division and design,to improve the efficiency of data query.Secondly,the SSM framework is used to complete the development of agricultural cultural tourism management platform.Spring MVC architecture is adopted to separate the model layer,view layer and controller layer,achieving the effect of high cohesion and low coupling.MyBatis persistence layer framework is used to encapsulate JDBC(Java Database Connectivity),so that the underlying operation of the Database becomes transparent.Write the script file of collaborative filtering recommendation algorithm in Python,and execute Java to execute the script file.Finally,the function,performance and platform page of the platform were tested,mainly including function completion and fault tolerance test.Through system testing and result analysis,the system security and application are ensured.