Sentiment analysis for points of interest : a point of interest recommendation system based on “W4” recommendation model

• Main Author: Li, Pengfei
• Other Authors: Cong Gao
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer applications
• Online Access: http://hdl.handle.net/10356/66825
• Institution: Nanyang Technological University

How to profile users based on Point of Interests (POI), and how to provide customized services to users have become practical social issues with huge potential commercial values. This project was to address the issues through implementing the “W4” POI recommendation model discussed in Yuan et al (2012), focusing on the four aspects of POI recommendation including the user (Who), the location and region (Where), the day and time (When) and the activity (What). This report discussed motivations of the project and various design process including scope finalization, project schedule and task assignments. The report further detailed concrete implementation considerations during each of the implementation phase, from preprocessing the raw data, constructing recommendation model, setting up prediction application server, designing interface, to integrating front-to-end communication of the system. The report also included recommendations for future implementations and enhancements. Lastly, the project delivered a web-based demonstration system with interactive user interface to provide the POI recommendation services.