Business location recommendation using check-in data

Abstract

In this era of Social Networks, large amount of data is generated from the users of social media and mobile applications on day to day basis. This data can be useful for the companies as they provide insight into the location oriented decisions of the businesses and on user behavior patterns in their regular activities. In this thesis work we are interested in the LBSN (Location Based Social network) data which is generated when the Users of social network Interact in the Online Social Networking Platforms and mobile applications by sharing their location data through “check ins” in the various Business locations. This spatial aspect of the LSBN data almost represent an online model of the physical world which can be analyzed to find key insights regarding the business locations. We have used the Geographical and Social distances to partition the city into neighborhoods as place for a new business opportunity. In technique we have used the collaborative neighborhood filtering based on similarity of neighborhoods by establishing correlation between business venues and check in patterns. We have used the New York foursquare data for our experimentation, this experimentation shows promising results for prediction of future business location.