BOTTARI: An Augmented Reality Mobile Application to Deliver Personalized and Location-Based Recommendations by Continuous Analysis of Social Media Streams

Abstract

In 2011, an average of three million tweets per day was posted in Seoul. Hundreds of thousands of tweets carry the live opinion of some tens of thousands of users about restaurants, bars, coffees and many other semi-public points of interest (POIs) in the city. Trusting this collective opinion to be a solid base for novel commercial and social services, we conceived BOTTARI: an augmented reality application that offers personalized and localized recommendation of POIs based on the temporally-weighted opinions of the social media community. In this paper, we present the design of BOTTARI, the potentialities of semantic technologies like inductive and deductive stream reasoning and the lesson learnt in experimentally deploying BOTTARI in Insadong – a popular tourist area in Seoul – for which we have been collecting tweets for three years to rate the few hundreds of restaurants in the district. The results of our study show to demonstrate the feasibility of BOTTARI and encourage its commercial spreading.

Keywords: Social Media Analysis, Mobile App, Personalized Recommendation, Location-Based Recommendation, Stream Reasoning