Insolvency is a complex issue, as the causes are diverse and the trajectory of a person experiencing insolvency varies.
Australian Government Financial Security Authority need to be equipped with additional capacity to predict the compliances of their potential clients and identify the best way to manage the debt collection processes on case-by-case basis.
Thus, capitalising on the powerful prediction capability of machine learning is vital to manage the complexity of insolvency process.
There is a need to utilise machine learning to analyse large volume of data that are collected from various government agencies on the same population. The machine learning model need to analyse the data at a personal level and to consider the external factors when make prediction to each client. Personalised prediction is needed as each client is unique in his/her circumstances that lead to and surrounding insolvency.
Our team built a complete machine learning model that harness large volume of government datasets from various organisations including: Australian Taxation Office, Australian Bureau of Statistics and Australian Government Financial Security Authority-non-compliance in personal insolvency data.
Visualising the output of our model is significant for decision and policy makers in government agencies to simplify large data and provide an insight to make the right informed decisions to the client and the society in large.