Mining Meta-Actions for Action Rules Reduction

Abstract

Meta-actions effect and selection are the fundamental core for a successful action rule execution. All atomic action terms on the left-hand side of an action rule have to be covered by well chosen meta-actions in order for it to be executed. The choice of meta-actions depends on the antecedent side of action rules; however, it also depends on their list of atomic actions that are outside of the action rule scope, seen as side effects. In this paper, we strive to minimize the side effects by decomposing the left-hand side of an action rule into executable action rules covered by a minimal number of meta-actions and resulting in a cascading effect. This process was tested and compared to original action rules. Experimental results show that side effects are diminished in comparison with the original meta-actions applied while keeping a good execution confidence. ding effect. This process was tested and compared to original action rules. Experimental results show that side effects are diminished in comparison with the original meta-actions applied while keeping a good execution confidence. between the objects' properties values transitions and the desired decision property values transition. This object property correlation results in a cascading effect that ultimately triggers a change in the decision value. Meta-actions are the triggers to those action rules and provide a tool to control their execution. They are represented by an influence matrix containing a set of attribute transitions they trigger. In most previous work, it is assumed that meta-actions are provided and ready to use. In addition, action rules are discovered and then chosen independently of the meta-actions. However, meta-actions are not commonly available in a ready to use format, and therefore they have to be mined and formatted. Action rules and their triggering meta-actions are interrelated and should be discovered together. In addition, meta-actions introduce negative side effects that might be harmful for the objects, and are not taken into consideration in the action rules discovery process. In other words, they provoke changes in objects attribute values that will not only trigger the actions rule targeted but also negative side effects. The negative side effects can damage the objects features outside of the executed action rule scope. Since the discovery process is bounded by the system in use, it is important to integrate the action rules and meta-actions discovery.In this paper we aim at connecting the action rules discovery process and meta action mining to reduce the negative side effects and improve the execution confidence of action rules. For this purpose, we propose a meta-action mining and evaluation process, and an action rules reduction method. The meta-action mining process is done through the study of expert knowledge treatments and their resulting effects. The action rules reduction connects the action rules to their meta-actions for a better action rule selection than the traditional action rule support and confidence se...