What makes patterns interesting in knowledge discovery systems

Abstract

One of the central problems in the eld of knowledge discovery is the development of good measures of interestingness of discovered patterns. Such measures of interestingness are divided into objective measures { those that depend only on the structure of a pattern and the underlying data used in the discovery process, and the subjective measures { those that also depend on the class of users who examine the pattern. The focus of this paper is on studying subjective measures of interestingness. These measures are classi ed into actionable and unexpected, and the relationship between them is examined. The unexpected measure of interestingness is de ned in terms of the belief system that the user has. Interestingness of a pattern is expressed in terms of how it a ects the belief system. The paper also discusses how this unexpected measure of interestingness can be used in the discovery process.

A generic robot database and its application in fault analysis and performance evaluation

“…The intent is to understand the emergence of Actionable Knowledge [11], that is Knowledge the robot can use to make decisions and act upon to achieve its tasks [12]. We classify content of the robotic mind into a hierarchy.…”

Measuring interestingness of discovered skewed patterns in data cubes

“…Form-dependent objective measures [4,5,7] such as neighborhoodbased unexpectedness, surprisingness and logical redundancy have been applied for ranking and clustering patterns such as association rules. Subjective interestingness measures require some form of user input in determining the utility of a mined pattern [18,21,27]. Utilitybased measures have been used for objective-oriented association mining (for example, [26,33]), with user-specified objectives.…”

Querying multiple sets of discovered rules

“…Rules can be utilized to gain a better understanding of the application domain, and to take actions to one's advantage. However, the number of rules generated from a dataset by some of the rule mining algorithms is often very large, in thousands or tens of thousands [4,13,15,16,17,20,21,24]. Making sense of such a large number of rules presents a significant challenge.…”

“…Making sense of such a large number of rules presents a significant challenge. Past research has shown that most of the rules are actually not useful or interesting for specific applications [15,16,17,20,21,24].…”