Researchers at the University of South Australia have developed a machine learning technique that makes property valuation more transparent, reliable and practical, with the ability to accurately model the impact of urban development decisions on real estate prices.
The technique was created and validated using over 30 years of historical sales information in the Adelaide metropolitan area and uses specially developed machine learning algorithms to process huge amounts of data on the housing, urban structure and equipment, making it possible to quantify the effects of urban planning policies. on the value of the home.
Lead researcher, UniSA geospatial data analyst and urban planning expert Dr Ali Soltani explains that the technique has implications for the real estate, urban planning and infrastructure sectors.
“Our modeling technique and results can help real estate investors, builders, owners, home appraisers and other stakeholders get a more realistic view of property value and the factors that affect it” , says Dr. Soltani.
“This research has implications for policy makers by providing insights into the potential impacts of urban planning – such as infill regeneration, planned communities, gentrification and population displacement – and the policies of providing infrastructure on the housing market and the resulting local and regional economy.
“By capturing the complicated influence of infrastructure elements such as road and transit networks, shopping malls, and natural landscapes on home value, our model is particularly useful in improving the accuracy of current forecasts of land values and reduce the risks associated with traditional property valuation methodologies, which are largely dependent on human experience and limited data.
Dr Soltani says the model – developed in collaboration with Professor Chris Pettit of UNSW’s City Futures Research Center – can also be extended to include other economic characteristics at macro and micro levels, such as changes in interest rates. interest, employment rates and the influence of COVID-19, harnessing the benefits of big data technologies.
“This model has the potential to be used as a decision support platform for a variety of stakeholders, including home buyers and sellers, banks and financial agents, investors, government and insurance or loan agents,” says Dr. Soltani.
“Our technique makes it easier for stakeholders and the general public to apply the results of sophisticated models on historical or real-time data from multiple sources, which were previously almost black boxes and expert-oriented.”
A summary of this research was recently published in the journal Cities by Elsevier.