Asthma is one of the most common long-term conditions in New Zealand (NZ), affecting 1 in 5 children and 1 in 8 adults. Asthma attacks are the leading cause of asthma deaths, placing a significant burden on our health system. Asthma attacks are however highly preventable if detected early. Unfortunately, there are currently no accurate, reliable ways of detecting asthma attacks before they occur. This project uses a data-driven approach to find out new knowledge of risk factors relating to asthma attacks by collecting real-time data from different data sources, such as data from 'smart' devices and wireless sensors, and combining with data from NZ datasets. We will apply artificial intelligence techniques to explore this data, so a more accurate and comprehensive risk prediction model can be developed for asthma attacks. This means attacks can be predicted more accurately and in real-time so health providers respond in a timely way.