A Bayesian Network Model for Interesting Itemsets
This work addresses the need for efficient and high-quality itemset mining in data analysis, though it appears incremental as it builds on existing interestingness models.
The paper tackles the problem of mining interesting itemsets for exploratory data analysis by proposing the first generative model over itemsets using a Bayesian network and a novel interestingness measure, demonstrating that it retrieves itemsets with quality comparable to or better than state-of-the-art algorithms on real-world datasets.
Mining itemsets that are the most interesting under a statistical model of the underlying data is a commonly used and well-studied technique for exploratory data analysis, with the most recent interestingness models exhibiting state of the art performance. Continuing this highly promising line of work, we propose the first, to the best of our knowledge, generative model over itemsets, in the form of a Bayesian network, and an associated novel measure of interestingness. Our model is able to efficiently infer interesting itemsets directly from the transaction database using structural EM, in which the E-step employs the greedy approximation to weighted set cover. Our approach is theoretically simple, straightforward to implement, trivially parallelizable and retrieves itemsets whose quality is comparable to, if not better than, existing state of the art algorithms as we demonstrate on several real-world datasets.