Restaurant Recommendation Algorithm Based On Category Preference And User Interest

Online ordering has gradually become a new form of dining,and when faced with countless online restaurant categories,reviews,ratings and other information,it is difficult for users to quickly and accurately select restaurants that they are interested in.In order to solve this problem,this dissertation proposes a restaurant recommendation algorithm based on category preference and user interest.First of all,a classification model based on the FTF-IDF algorithm and Word2Vec is established,and the restaurant review text is mainly divided into three categories(taste,environment,and service)according to users' personal preferences.Secondly,a recommendation algorithm based on user multi-attribute and user-restaurant rating similarity is proposed to make personalized restaurant recommendation for users.The main research contents are as follows:(1)Aiming at the problem of inaccurate meal scores.This dissertation establishes a classification model based on FTF-IDF algorithm and Word2Vec.First of all,considering that TF-IDF does not consider the influence of the word's part of speech on the calculated value when calculating the frequency of the text,it is proposed to introduce a contribution factor on the basis of the traditional TF-IDF algorithm.Secondly,by adding the contribution factor in TF-IDF to obtain the weight value based on the part of speech,and use it as the weight value of the word vector of the Word2Vec model,and then calculate the word vector of each document;finally,use the SVM support vector machine for training,get classified experimental text.(2)Aiming at the problem of low recommendation accuracy.This dissertation constructs a recommendation algorithm based on user multi-attribute and user-restaurant rating similarity.First,calculate the attribute similarity between users based on their basic information.The main information includes gender,age,and occupation.Secondly,construct a user-restaurant rating matrix,propose a restaurant rating formula that comprehensively considers the different preferences of users,and calculate the similarity of restaurant ratings between users.Finally,the two similarities are linearly weighted,the score prediction formula is used to predict the score,and the top-N is selected to recommend restaurants to users.The experimental results show that the restaurant recommendation algorithm based on category preference and user interest proposed in this dissertation can effectively and accurately improve the accuracy of restaurant classification,user dining experience and restaurant recommendation accuracy.This dissertation innovatively considers the user's preference for restaurant categories and the influence of user attribute information on the restaurant recommendation system,and uses text classification and constructing a scoring formula to predict the score.The research in this article can effectively classify restaurant reviews;at the same time,building a user-restaurant rating formula by using a weight function can effectively make personalized restaurant recommendations for users.In addition,the research in this article also provides a theoretical basis for the follow-up development of the catering industry.Figure[30]table[14]reference[63]...