Mining Social Media photos to Measure Point-of-Interest Connectivity

Abstract

The capability of social media to allow users to upload their geo-tagged and text tagged photos has led to the creation of a voluminous data, which can be mined to extract useful information. Most of these photos are public and can be accessed and analyzed for extracting useful patterns. This paper presents a prediction of Point of Interest (POI) connectivity through mining these publically available photos on social media. Two POI are considered connected for a particular user, if the user has taken photos of the two points, within certain duration of time. The proposed research can be utilized in (1) building a recommender system that would help the tourists planning their travel sequence and (2) helping the tourism management authorities to predict the tourism statistics of the nearby POI once they have collected statistics of a particular POI. This information can be utilized in efficient resource management for sustainable tourism.