Unexpectedness as a Measure of Interestingness in Knowledge Discovery

Organizations are taking advantage of quot;data-miningquot; techniques to leverage the vast amounts ofdata captured as they process routine transactions. Data-mining is the process of discoveringhidden structure or patterns in data. However several of the pattern discovery methods in dataminingsystems have the drawbacks that they discover too many obvious or irrelevant patternsand that they do not leverage to a full extent valuable prior domain knowledge that managershave. This research addresses these drawbacks by developing ways to generate interestingpatterns by incorporating managers' prior knowledge in the process of searching for patterns indata. Specifically we focus on providing methods that generate unexpected patterns with respectto managerial intuition by eliciting managers' beliefs about the domain and using these beliefs toseed the search for unexpected patterns in data. Our approach should lead to the development ofdecision support systems that provide managers with more relevant patterns from data and aid ineffective decision making