Data mining in law with association rules

We apply the data mining technique called association rules to data from family law in Australia in order to demonstrate that this can be an invaluable tool for legal analysts. The rules identify associations between variables and are discovered automatically. Interesting rules can prompt analysts to formulate hypothesis for further investigation. In general, data mining techniques have not been applied to law despite benefits such as those advanced by association rules. This is because data that represents decision making processes in law are typically recorded in judgements as narrative and not in a more structured way such as in a database. We argue that if knowledge is represented as arguments within decision support systems, then data can be collected systematically while facilitating the process of drafting narratives.