Data-driven rental price analysis

• Main Author: Tey, Kai Seong
• Other Authors: Sourav S Bhowmick
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Data-driven; Rental price; Data engineering
• Online Access: https://hdl.handle.net/10356/181144
• Institution: Nanyang Technological University

In this study, we developed a data-driven rental price analysis platform for Singapore's rental market. The project implemented an end-to-end solution encompassing automated data collection, processing, analysis, and visualisation. Web scrapers collected data from multiple online rental marketplaces, which was then processed and stored in a cloud-based OLAP database. Machine learning models, particularly Random Forest algorithms, were employed to analyse patterns and predict rental prices with 95% accuracy. The system architecture utilised Apache Airflow for workflow orchestration and Streamlit for dashboard development. Deployed on-premise using Kubernetes, the platform provides stakeholders with valuable insights into rental market dynamics. Despite limitations in real-time processing and dataset comprehensiveness, the study successfully delivered an automated tool for analysing Singapore's dynamic real estate market, offering a foundation for future enhancements in data collection and macroeconomic integration.