Development of point-of-interest recommendation web application

• Main Author: How, Jiahe
• Other Authors: Gao Cong
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering::Computer science and engineering::Information systems::Information systems applications
• Online Access: http://hdl.handle.net/10356/67399
• Institution: Nanyang Technological University

The use of recommendation based systems has been a growing trend in the last decade. In today’s modern technology age, whereby users have access to high bandwidth network, it has been a norm for people to frequently visit recommendation website to look for point-of-interests such as nearby restaurants or tourist attractions. However, most recommendation systems such as “HungryGoWhere” would only provide recommendation for 1 type of category, such as food for this instance, which proves to be uninformative for general users. The purpose of the project is to combine ideas from various recommendation system available in the market and create a web application that would provide recommendation for a range of categories with enhanced features. Firstly, data crawlers was created to crawl venues and tips data from Foursquare which is used as the database for the web application. The setting of the crawler allow it to crawl venues from any countries specified. Secondly, a web application that function as a recommendation system was built which provides the recommendation of point-of-interest for a ranges of categories. Additional feature such as Google Street View and reviews sections of each venue was added to enhance user experience. The web application structure was designed for high maintainability as well as loose coupling whereby each features and layers are distinctly separated that enable limited efforts for future improvement work. By the end of the project, all objectives were met and accomplished. The crawler has crawled data for both Singapore and London and comparison were made between them. The web application were able to recommend places based on user preferences and contains features that enhance its usability. Further analyzing of the crawled data and user’s behavior could give in-depth view of the venue density of an area as well as future improvement for the User Interface of the web application.