In many business contexts, the ultimate goal of knowledge discovery is not the knowledge itself, but putting it to use. Models or patterns found by data mining methods often require further post-processing to bring this about. Actions and Action rules are two types of knowledge thatsuggest some changes in object properties to gain a profit in corresponding domain. . Currently, action mining methods rely on predictive models, obtained through data mining, to estimate the effectof certain actions and finally suggest actions with desirable effects. A major problem with this approach is that predictive models do not necessarily reflect a causal relationship between their inputs and outputs. This makes the existing action mining methods less reliable. In our dissertation, we've introduced action mining based on causal structures. First, we've proposed CREAM, which can learn actions from a Causal Network, then ICE-CREAM, a novel approach to action mining that explicitly relies on an automatically obtained best estimate of the causal relationships in the data, and finally CARE, an action rule extracting method using Causal Networks.Experiments confirm that proposed methods perform much better than the current state of the art in action mining. Key Words Action Mining, Action Rules, Causality, Causal Networks, Causal Rules.
