A Food Venue Recommender System Based on Multilingual Geo-Tagged Tweet Analysis

This paper proposes a novel system which utilizes information from a social network services to suggest food venues to users based on crowd preferences. To recommend an appropriate food venue for each crowd preference, the system ranks food venues in each region by using an improved collaborative filtering method based on the differences between locations and languages in geo-tagged tweets. A key feature of the proposed system is the ability to suggest food venues in regions where very few geo-tagged tweets are available in a specific language by using the weighted similarity by others' preferences. To implement the system, more than 26 million tweets from European countries were collected and analyzed based on 6 languages and 7 regions. Afterwards, we provide an evaluation of the ranked venues proposed by the system based on 89 French speakers in 7 European countries.