Sentiment analysis of property data

• Main Author: Lau, Sze Ye
• Other Authors: Ong Yew Soon
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Document and text processing
• Online Access: http://hdl.handle.net/10356/62829
• Institution: Nanyang Technological University

In this new age of information, the Internet holds a large amount of data which is increasing every minute. This is partly due to the social media being the latest fad where a large amount of people express themselves online. These sentiment data are extremely attractive to both business and consumers as useful knowledge can be obtained after analysing it. Now consider the property market where it has always been a blooming industry but suffers from a lack of innovation. With the recent cooling measures the government has implemented, in order to stay competitive and relevant, new solution needs to be implemented to appeal to customers. In this project, we aim to innovate the property market in Singapore by utilising sentiments given by consumers on the service of the property agents and the properties. With the current generation being more tech-savvy, most consumers tend to seek information online first before making any purchase decision. Instead of relying on traditional media to disseminate information, a web portal will be created for property agents to list their properties. This allows potential buyers to search and browse through a large amount of information with ease. There will also be a map interface to provide better visual aid for customer. There will be a review section for customers to comment on the agent’s service and on the properties. These reviews serve as additional information for other buyers as well as an indirect feedback channel to the agents and their company. The project will include a web portal and a sentiment analyser for analysing the reviews. Web crawler and parsers are utilised in this project to obtain real world data which is required for the training for the sentiment analyser.