Object-Driven Action Rules

Object-Driven Action Rules

Ayman Hajja (University of North Carolina, USA) and Zbigniew W. Ras (University of North Carolina, USA & Warsaw University of Technology, Poland)
Copyright: © 2015 |Pages: 10
DOI: 10.4018/978-1-4666-5888-2.ch114

Chapter Preview


Object-Driven Action Rules

In this section, we start by providing the reader with the necessary background to shape a complete picture behind object-driven action rules and its motivation. Classical action rules are introduced, formally defined and fully explained, to the extent of preparing the reader with the necessary background to capture its relation to object-driven action rules, by the similarities of both, and the novelty of each.

Following the background, we move our discussion to object-driven action rules; starting by clearly communicating to the reader the properties of datasets that work for action rules concepts and by fully explaining the assumptions we make for our adapted system.

Key Terms in this Chapter

Atomic Action Set: An expression that describes a valid state transition for a single attribute.

Action Rules: Rules extracted from a decision system that describe a possible transition of objects from one state to another with respect to a distinguished attribute called the decision attribute.

Confidence: The probability for an action set or an action rule to be true.

Support: The number of instances that satisfies an action set or an action rule.

Czermak’s Mirror Test: A test performed by medical doctors to measure the amount of nasal air escape of the patient. It is used to indicate the condition severity of hypernasality patients.

Complete Chapter List

Search this Book: