Information fusion in location-based services

Abstract

Location-Based Social Networks(LBSNs) have emerged as major sources of information about users as well as locations. Different LBSNs have different ratings and reviews for a particular location and each claims to have accurate and genuine information. This creates a need to nd the credibility of a location-based on its user rated attributes such as rating, reviews, etc. Also, latitude-longitudes of the same locations on different networks are not usually identical (because of nearby places can have very close coordinates) and this poses a problem in matching these locations on different networks, and also raises questions on the credibility of the location. In this project, our aim is to match the same locations on different networks using the location attributes and also to determine the anomalies in the user rated attributes of the location for its credibility check. We experimented over 50,000 unique locations from 30 cities obtained from Foursquare, Yelp, TripAdvisor and Zomato.