Research On Context-aware Tourism Service Resource Recommendation System

With the advent of the information explosion era,the information on tourism service resources has also grown explosively.However,it is quite difficult for users to obtain tourism service resource information in a timely,efficient,and accurate manner.In order to improve users’ access to and utilization of tourism service resources,the industry has designed and developed tourism service resource recommendation systems.The system can analyze users and tourism service resources,and provide personalized recommendations of tourism service resources to users.However,due to the close correlation between tourism service resources and users’ current situations,when users’ situations change,their utilization of tourism service resources will change as well,thus affecting the recommendation effect of the system.Currently,many tourism service resource recommendation systems do not fully utilize users’ situational information.To further improve users’ access to and utilization of tourism service resources,this study proposes a context-aware framework for tourism service resource recommendation.The framework is divided into multiple levels,and this study analyzes and designs each module in detail at each level,especially designing and implementing algorithms for some modules in the context modeling and recognition layer and the recommendation layer.At the same time,this study evaluates the overall recommendation effect of the context-aware collaborative filtering algorithm,and the results show that it can achieve better recommendation effects compared to a single collaborative filtering algorithm.Finally,this study analyzes and designs a prototype system of a context-aware tourism service resource system,and designs and presents the front-end and back-end pages for both users and administrators.Innovation:(1)This article proposes a contextual-aware tourism service resource recommendation framework that can model contextual information and blend multiple traditional recommendation algorithms to improve recommendation accuracy.(2)In terms of algorithm effectiveness evaluation,a filtering threshold is set based on the size of contextual similarity,and the impact of threshold values on the degree of improvement of traditional collaborative filtering algorithms is studied.