Design And Implementation Of Travel Recommendation System Based On Improved Collaborative Filtering Algorithm

With the deepening of the integration of the tourism industry and the Internet,the Internet of tourism information has become an inevitable trend.The huge amount of users and tourism information data makes the problem of information overload increasingly serious.In order to change the current situation of users searching for tourist attractions from massive tourist information,this thesis applies collaborative filtering algorithm to travel recommendation to provide users with accurate travel recommendation services.However,because the traditional collaborative filtering algorithm only relies on a single user rating data and has its own shortcomings that cannot meet the complex needs of users in the travel industry,this article improves the traditional collaborative filtering algorithm and designs and implements travel recommendation on this basis system.In travel recommendation,in addition to user rating data,the characteristics of the attractions themselves are also factors that cannot be ignored,so this article integrates the attributes of the attractions into the collaborative filtering recommendation algorithm.First,the tourist attractions are divided into nine categories,and the crawled attractions profile data are manually labeled and classified.Establish an evaluation index system for tourist attractions that can specifically express the characteristics of the attractions.After crawling the relevant index data of each attraction,the data is preprocessed,and the standardized index data is weighted to calculate the similarity between the attractions to obtain a list of attractions similarity.Then,a user-attraction rating matrix is established based on the obtained user rating data,and the similarity between users is calculated and sorted to obtain a pre-recommended list.Finally,compare the importance of the attractions in the similarity list and the pre-recommended list,sort the results and select TOP-N to generate the final recommendation list.The improved algorithm model is applied to the design and development of the tourism recommendation system to realize the recommendation of tourist attractions in 11 cities in Hebei Province.In this thesis,we innovate the evaluation index of scenic spots and design indicators that meet the recommendation of scenic spots.An improved collaborative filtering algorithm model is proposed.Experiments show that this model has moreaccurate and comprehensive recommendation results,and a travel recommendation system based on the improved algorithm model is designed and implemented.