A vast amount of information is provided through the Internet. This makes it difficult for you to sift through useful information for yourself. In other words, uncertainty and ambiguity are increasing in information selection. To solve this problem, many researchers have recently become interested in recommendation services among personalization tools. In order to provide recommendation services to users, recommendation systems and interfaces must be developed, But it is biased as a recommender system.
However, improvement in the accuracy and efficiency of the recommendation system does not necessarily lead to the decision-making and satisfaction of the recommendation service users. These problems and technical limitations were raised. A fundamental limitation that cannot be resolved is that recommendations can only be made within the range of data existing in the database. In other words, even if a very sophisticated model is developed, it is bound to be a limited recommendation.
What can compensate for this is to personalize the user interface of the recommendation site that connects the user and the recommendation system. Therefore, the need to expand the scope of research to user experience that encompasses various aspects of users, away from research focused on recommendation systems, has been raised, and the scope of research is expanding.
In order to improve user convenience and user satisfaction while supplementing the fundamental limitations of the recommendation system by developing a personalized user interface, the attributes considered important when a user selects a restaurant, that is, the restaurant selection attribute, must be reflected.
Accordingly, this study derived explicit attributes from restaurant recommendation sites, portal sites, and review sites. The derived attributes were divided into text information and photo information. Based on the attributes of the classified information, the social network analysis methodology was applied to analyze the restaurant selection attribute as a relational characteristic between the attributes included in each information. In addition, differences in selection attributes according to demographic characteristics were analyzed.
The implications derived from this study are as follows. First, it is necessary to make efforts to revitalize restaurant recommendation sites and the Korea Tourism Organization website. Second, the provided text information of restaurants should be changed. Third, it is necessary to change events using SNS. Fourth, the photo information of the restaurants provided should be unified. Fifth, a personalized user interface needs to be developed.
In order to revitalize the recommended service, the reliability of the recommended service should be enhanced and continuous promotion by developing a personalized recommendation system and developing a personalized user interface that reflects selection attributes.