Discovering Action Rules That Are Highly Achievable from Massive Data
PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining
Action rule extraction from a decision table: ARED
ISMIS'08 Proceedings of the 17th international conference on Foundations of intelligent systems
Discovering the concise set of actionable patterns
ISMIS'08 Proceedings of the 17th international conference on Foundations of intelligent systems
Combined association rule mining
PAKDD'08 Proceedings of the 12th Pacific-Asia conference on Advances in knowledge discovery and data mining
ISMIS'11 Proceedings of the 19th international conference on Foundations of intelligent systems
An expected utility-based approach for mining action rules
Proceedings of the ACM SIGKDD Workshop on Intelligence and Security Informatics
Mining actionable behavioral rules
Decision Support Systems
| Hi-index | 0.00 |
There are two aspects of interestingness of rules, objective and subjective measures ([7], [1], [15], [16]). Objective measures are data-driven and domain-lly, they evaluatethe rules based on their quality and similarity between them.Subjective measures are user-driven, domain-dependent, and include unexpectedness, novelty and actionability. 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 by authors that actionability has to be expressed in terms of changes in values of certain attributes which are used in an informationindependent. Genera system. They introduced a new class of rules (called action rules) which are constructed from certain pairs of classificationrules extracted from the same information system. Conceptually similar definition of an action rule was proposed independently in [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 classi?cation part is either the shortest