BLOSOM: a framework for mining arbitrary boolean expressions

We introduce a novel framework (BLOSOM) for mining (frequent) boolean expressions over binary-valued datasets. We organize the space of boolean expressions into four categories: pure conjunctions, pure disjunctions, conjunction of disjunctions, and disjunction of conjunctions. For each category, we propose a closure operator that naturally leads to the concept of a closed boolean expression. The closed expressions and their minimal generators give the most specific and most general boolean expressions that are satisfied by their corresponding object set. Further, the closed/minimal generator expressions form a lossless representation of all possible boolean expressions. BLOSOM efficiently mines several forms of frequent boolean expressions by utilizing a number of methodical pruning techniques. Experiments showcase the behavior of BLOSOM with respect to different input settings and parameter thresholds. An application study on real datasets is also given. This work was supported in p...

• Type: Conference
• Year: 2006
• Where: KDD
• Authors: Lizhuang Zhao, Mohammed J. Zaki, Naren Ramakrishnan
• Keywords: Boolean Expressions, Data Mining, Frequent Boolean Expressions, KDD 2006, Possible Boolean Expressions