Singapore Neighbourhood Recommendation System

Elizabeth Stepton, Joshua Vargas, Ning Xinran, Sean Lim

Data Science, Software Development, Product Management, HCI

Project Brief

Singapore Neighbourhood Recommendation System is a final project for YSC2244 Programming for Data Science, a course in the Data Science track in the Yale-NUS Mathematical, Computational, and Statistical Sciences programme. The aim of this project is to deploy techniques learned in YSC2244, as well as self-taught technologies, to address a real-world issue that can be addressed through a data-driven application.

Target Audience

Singapore is one of the world's most expensive property markets. Prospective homebuyers are faced with information asymmetry, and it is difficult to narrow one's property search. This inspired us, a group of Yale-NUS students (including three seniors and two internationals), to develop a prototype for a tool we would have loved to use through utilising open-source GIS solutions and publicly-available data.

Skills Utilised

DATA SCIENCE

Geospatial data science

Data Scraping

Data Filtering Algorithm Design

Interactive Data Visualisation

SOFTWARE DEVELOPMENT

Product Design

User Stories

Specification Writing

UI/UX Design

Version Control (GitHub)

GRAPHIC DESIGN

UI/UX Design

Singapore is now the 5th most expensive city in the world for property prices, according to the 2022 Knight Frank Wealth Report. It is closing the gap with 4th-placer New York and is already ahead of Shanghai, Paris, and Tokyo. Amid a general global downturn, property prices are soaring in Singapore, leading the government to announce unprecedented property cooling measures to prevent the housing market from becoming a bubble.

One oft-cited reason for the rise in property prices and rent has been information asymmetry. Tenants and homebuyers are left at a disadvantage compared to real estate agencies and landlords, allowing the latter to charge higher rates (a ‘resetting’ of market rates).

Our project aims to help address this information asymmetry by giving prospective homebuyers insights into neighbourhoods so that they can narrow down their choices. Our team implemented a website for prospective homebuyers to identify the most suitable subzone zones in Singapore based on their priorities and preferences. Our final deliverable was an interactive website that allows the user to indicate their preferences via a series of sliders, and the results are generated via a filtered database and a map that visualizes these results.

Future extensions could make the project useful for renters; enable real-time property price scraping and prediction; and take into account other factors such as BTO launches for Singaporean citizens and PRs.

Try App

This opens an external website.

Singapore Neighbourhood Recommendation System

Where to Find Me