Discovery of Interesting Action Rules

There are two aspects of interestingness of rules, objective and subjective measures ([7], [1], [15], [16]. Objective measures are datadriven and domain-independent. Generally, they evaluate the rules based on their quality and similarity between them. Subjective measures are user-driven, domain-dependent, and include unexpectedness, novelty and actionability [7], [1], [15], [16]. Liu [7] defines a rule as actionable one, if user can do an action to his/her advantage based on that rule. In [12] it was assumed that actionability has to be expressed in terms of changes in values of certain attributes which are used in an information system. They introduced a new class of rules (called action rules) which are constructed from certain pairs of association rules extracted from the same information system. Conceptually similar definition of an action rule was proposed independently by [4]. Action rules have been investigated further in [14], [13], [11], and [18]. In order to construct action rules it is required that attributes in a database are divided into two groups: stable and flexible. Flexible attributes are used in a decision rule as a tool for making hints to a user what changes within some of their values are needed to reclassify a group of objects from one decision class into another one. In this paper, we give a strategy for constructing all action rules from a given information system and show that action rules constructed by system DEAR, presented in [13], cover only a small part of all action rules. Clearly, we are not interested in all action rules as we are not interested in extracting all possible rules from an information system. Classical strategies like See5, LERS, CART , Rosetta, Weka are discovering rules which classification part is either the shortest or close to the shortest. This approach is basically ruling out all other classification rules unless their are surprising rules [17]. In this paper, we introduce the notion of a cost of an action rule and define interesting action rules as rules of the smallest cost. We give a strategy showing how interesting action rules can be generated from action rules discovered system DEAR.