Research On Tourism Recommendation Based On Online Review

The rapid development of tourism Internet applications has produced a large number of reviews related to tourist attractions.These comments reflect the thoughts and preferences of tourists on field trips or services after traveling on the ground,and appear in various forms on blogs,BBS or forum websites and other media,becoming a vehicle for more and more important experience information,and potential tourists also take a lot of time to read online reviews to assist with travel decisions.However,the commentary information written by tourists is arbitrary,and the value of the commentary content varies greatly.The large amount of redundant information seriously interferes with the potential tourists to obtain and analyze their useful information.Information overload problem makes it difficult for potential tourists to make appropriate personalized travel decisions.Therefore,it is necessary to study from the mass of online commentary text content,to tap the experience knowledge contributed by the many people,and then combine the concerns of individual travel preferences,so that when tourist attractions are selected,they can effectively use group wisdom to formulate more reasonable and individualized travel decisions.The tourist activities involve six aspects of eating,traveling,traveling,purchasing,housing,and entertainment.Each aspect has a large amount of data.These data include not only text and images,but also spatial locations,travel routes,and other geospatially-related attributes.Therefore,how to use geographic information system to collect,store,manage,calculate,analyze,display,and describe tourism data that contains a large amount of information with spatial characteristics,and provide visitors with rich tourist experience and personalized tourism services has became an inevitable requirement for the sustainable development of tourism in the future.This article is based on the reading of a large number of existing related documents.The main research contents and results are as follows:(1)Based on the commentary data published after the traveler's field trip,a mathematical model of attraction recommendation index was built to provide tourists with personalized attraction recommendations.Firstly,according to the information content of the online travel commentary,on the one hand,the review data of the to-be-recommended scenic spot is selected,a fine-grained sentiment analysis is performed on the review object,and on the other hand,the historical review data of the tourist site to be recommended is selected,and the focus of attention commentary object is extracted by word frequency statistics.Then,building attractions recommendation index based on personal preferences and community wisdom,and finally the attractions with high attractions recommendation index are recommended to tourists.(2)A tag based tourism recommendation method considering emotion analysis is proposed.First,crawling the content information of tourist reviews on the website,and converting the data into the desired description of the scenic spots.At the same time,analyzing the review document or description before the target visitors,thus extracting the tourist's personal interest based on the content information that the tourists have commented on.Then,the description tag and the interest label of the tourist picture are set up respectively.Finally,by comparing the performance of the tag based recommendation algorithm before and after the emotional analysis,it is found that the algorithm after the emotional analysis has a high accuracy.(3)Focusing on the business needs of tourist travel,focusing on the content recommended by tourists' personalized attractions,and drawing on the open source lightweight Struts2,Spring,Hibernate framework(or SSH)integration strategy,applying the recommended method proposed in this paper as the recommendation engine,based on the combination of GIS technology and personalized recommendation technology,achieving a GIS-based travel recommendation system.The user functions of the recommendation system mainly include popular recommendation,recommendation index,tag-based recommendation,route planning,etc.,providing tourists with intelligent travel information services.