Tourism recommendation based on vector space model using composite social media extraction

Intentionally or not, social media users likely to share others recommendation about things, included tourism activities. In this paper we proposed a technique which was able to structure the joint recommendation of composite social media and extract them into knowledge about the tourist sites by deploying the vector space model. We included advice seeking technique to not only calculate recommendations obtained from the profile itself but also recommendations by social network users. This is a potential solution to handle sparsity problem that usually appears in conventional recommender systems. We further formulated an approach to normalize the unstructured text data of social media to obtain appropriate recommendation. We experimented the real world data from various source of social media in R language. We evaluated our result with Spearman's rank correlation and showed that our formulation has diversity recommendation with positive correlation to user's profile.