The Design And Implementation Of Recommendation System For Catering Based On Sentiment Analysis

With the continuous development of Internet technology,the emerging of "Internet+" has made the relationship between information technology and traditional service industry more and more closely recently.However,the problem of information overload between massive commodities and natural human users has always existed,and the recommendation system based on different types of social networks focus on how to solve this problem.Based on a recommendation model combined with user evaluation sentiment analysis,a recommendation system for the catering industry based on spark data platform has been designed and implemented.This system includes requirement analysis,outline design,detailed design and code implementation.The main work is as follows:1.A product recommendation model combining with dichotomous graph and sentiment analysis is proposed.Combining with user attribute information,user behavior information and commodity information,we put forward the rough sorting of the selected commodity set by using the binary graph recommendation algorithm.At the same time,sentiment analysis algorithm based on the features of sentiment dictionary calculate the comprehensive score of commodity evaluation.Combined with the score,the rough sorted goods are filtered again,and the final recommendation list of goods is formed.It shows that this method is more effective than the traditional recommendation algorithm.2.A recommendation system for the catering industry based on sentiment analysis is designed and implemented.This system uses B/S architecture,Spring Boot framework,Vue.js and other Java Web related technologies,The front and rear ends are developed and achieved.The functions of the whole system include commodity purchase,commodity evaluation,commodity recommendation functional modules.The product recommendation function is coordinated with Spark platform,and the system can make full use of the real-time processing capability of Spark Streaming to realize the product recommendation function efficiently.