Research On Recommendation Algorithm Of Tourist Attractions Based On User Clustering

The rapid development of the Internet brings us a lot of information and convenience.In the meantime,it also causes problem of excessive information,which leads us to filter information and invent the recommender system.The recommender system can meet individual need of users and improve the efficiency of access to useful information.Therefore,more and more relative researches about the system have been studied,and the system has been applied to varied fields,including the choice of tourist attraction.Generally,the Collaborative Filtering Recommendation is usually recommended,while it has restrictions to fully applied to the recommendation of tourist attraction,and it needs improvement to meet tourists' personal requirements.This article introduces relative researches about tourism recommendation and analyzes the significance of the recommender system.This article also introduces the history and background of the recommender system and concludes the advantages and-disadvantages of varied recommendation algorithm,and choses User-based Collaborative Filtering Based on the Clustering by comparison to conduct the studies of individuation of tourist attraction.However,the Collaborative Filtering Recommendation has the problem of data sparsity and cold start,so it should be improved.This article choses the way of clustering and conducts the study through experiments,and takes advantage of the K-means of clustering to collect the users of the data sets,and inputs the result of the clustering into Collaborative Filtering Recommendation to acquire the recommended result and compares it with the traditional Collaborative Filtering Recommendation.According to experimental results,the MAE and cost time of User-based Collaborative Filtering Based on the Clustering is less than that of traditional Collaborative Filtering Recommendation,which proves it can adopt the tourist and also proves that User-based Collaborative Filtering Based on the Clustering is more accurate-and rapid.This article improves the User-based Collaborative Filtering Based on the Clustering in the recommendation system of tourist attraction and increases the MAE and saves the cost time of the system by using the K-means of clustering,which will make a difference to the personal recommendation of tourist attraction.