Construction Of Rental Recommendation Model And Implementation Based On WeChat Mini Program

The era of big data has arrived with the continuous development of information technology and the Internet,huge data resources have become important assets in many industries.In many fields such as e-commerce,personalized advertising,social networking,logistics and distribution,video and audio,big data is assisting enterprises to continuously innovate operating models and develop new businesses.However,information overload also makes enterprises and users face great challenges.It is very difficult for users to find valuable and interesting information in the massive data,for enterprises,it is also necessary to understand the needs of users and provide users with convenient and efficient services and high-quality content.Recommendation system is the tools to solve this problem.The recommendation algorithm finds the user's personalized needs by analyzing the user's behavior,and then filters out the content that the user may be interested in from the massive data and pushes it to the user.Different business scenarios have different characteristics.Only by optimizing the recommendation algorithm in combination with the actual situation can it give efficient and accurate recommendation results.This paper proposes the construction of a rental recommendation model and the realization of a We Chat applet based on the rental scenario structure.The main contents are as follows:First,propose a rental model based on label weights.Combining online real-time recommendation,offline recommendation,feedback recommendation and other recommendation methods,it provides a recommendation algorithm with high operating efficiency,accurate results and strong interpretability.Using singular value decomposition and matrix dimensionality reduction to analyze and calculate all user behavior data,extract the most interesting listing labels for all users,and solve the cold start problem in the recommendation algorithm.Secondly,propose a density peak-based k-medoids clustering algorithm for location-based recommendation.Compared with the traditional k-medoids clustering algorithm,the k-medoids clustering algorithm based on density peaks has higher operating efficiency and stability.It makes up for the shortcomings of the rental model based on the label weight,and determines the area where the user wants to find a house in the rental scene through the discretely distributed longitude and latitude.Finally,combined with the rental recommendation model,a rental system based on We Chat miniprogram is designed and implemented.The system is highly portable,scalable and easy to deploy.The functionality of the system can be expanded according to specific needs,and has strong practicality.