Group Travel Recommendation Research Based On User Generated Spatial Data

In recent years,people's work and life have been greatly facilitated by the rapid development of emerging technologies like cloud computing.Internet of Things,mobile Internet artificial intelligence and so on.In terms of tourism and leisure,users can easily search for travel information through the Internet,purchase travel products and services,and enjoy the convenience of information technology.However,when faced with explosive growth of network information,it is difficult for users to make efficient choices.Therefore,the recommendation system in the past two decades has become a key issue in academic research,resulting in many methods and techniques.These recommendation system research results are increasingly popular among practitioners and are widely used including movies,news,books,restaurants,clothing,financial services,insurance,social labels,and general products.The recommendation system has made many breakthroughs in the vertical field.It has achieved good results in the recommendation of news and web pages and the recommendation of traditional products such as books and movies,but there are still many challenges when applying it to travel recommendations.As the demand for tourism continues to increase,the way people travel is more and more diverse.The Internet has greatly facilitated the travel of people,and has solved the problem of information asymmetry caused by traditional travel to a certain extent.The major travel websites collect the interest point information related to the purchase and purchase of the travel and purchase for the user to make a reservation,and the user can plan their own itinerary and route in advance.The recommendation system for tourism has also emerged.Many scenic spots have many classifications.How to recommend their preferred attractions for multiple users is a problem that the recommendation system needs to solve.By recommending the system,the cost of the planned itinerary is reduced for the user,and the travel preparation time is reduced,which is a problem that the group travel recommendation needs to solve.This paper summarizes the concept and characteristics of personalized tourism by studying the related theories of personalized tourism and the related literatures of group recommendation.By studying the related algorithms of personalized recommendation,the algorithm of user-based collaborative filtering is improved.Based on the spatial data of the user's uploaded travel photos,the unsupervised clustering method(DBSCAN)is used to fit the scenic spots,and the user's scenic spot access set is obtained.By combining the number of visits by the user at the attraction,the number of uploaded videos is uploaded.The number of photos obtains the user's rating on the attraction,and the popular attraction penalty factor is introduced to improve the similarity algorithm between users,and to better discover the attraction data of the spatial data set that meets the requirements of the personalized attraction.Based on the key techniques recommended by the group,a group recommendation fusion model based on travel experience weights is proposed based on the weighted fusion model and the recommended fusion method.The group recommendation fusion is performed for the personalized recommendation result through the travel experience-based weighted recommendation result fusion model proposed in this paper.Finally,it is proved by experiments that the fusion strategy based on travel experience can be better applied to the preference fusion of tourist attractions in group travel than the traditional fusion strategy.