Exploiting Geographical Data to Improve Recommender Systems for Business Opportunities in Urban Areas

The rapid urban expansion of the world's major cities has directly impacted people's lives. In the urbanization process, it is common that business shops are open to attend the different needs and demands of the increasing number of citizens. This fact represents a business issue encouraging potential investments that could be harnessed to improve both urban economic environment and quality of urban life. However, many business opportunities are lost or not exploited properly due to the difficulty that investors, business owners, and marketers have to identify the right places where to open new stores. In this paper, we describe the implementation and evaluation of an approach to identify geographic areas with great potential to host business from a specific category. First, we adapt clustering algorithms to work with geographical data and, thus, partitioning a target city into business districts. Next, we use various recommendation algorithms to suggest the best categories for each business district. We conduct several experiments on Yelp data and our results show how geographical data and state-of-the-art algorithms can be used to mine business opportunities and predict adequate places to open new stores in urban areas.