Using interesting sequences to interactively build Hidden Markov Models.

Scalable pattern mining with Bayesian networks as background knowledge.

Minimum Variance Associations %2D Discovering Relationships in Numerical Data.

Schema matching on streams with accuracy guarantees.

Efficient AUC Optimization for Classification.

Approximating Representations for Large Numerical Databases.

Comparison of Information Theoretical Measures for Reduct Finding.

Polynomial association rules with applications to logistic regression.

Mining rank%2Dcorrelated sets of numerical attributes.
We study the mining of interesting patterns in the presence of numerical attributes. Instead of the usual discretization methods, we propose the use of rank based measures to score the similarity of sets of numerical attributes. New support measures for numerical data are introduced, based on extensions of Kendall’s tau, and Spearman’s Footrule and rho. We show how these support measures are related. Furthermore, we introduce a novel type of pattern combining numerical and categorical attributes. We give efficient algorithms to find all frequent patterns for the proposed support measures, and evaluate their performance on real-life datasets.
Generalized Conditional Entropy and a Metric Splitting Criterion for Decision Trees.

A new metric splitting criterion for decision trees.

Finding Reducts Without Building the Discernibility Matrix.

Fast discovery of unexpected patterns in data%2C relative to a Bayesian network.

Interestingness of frequent itemsets using Bayesian networks as background knowledge.

A Metric Approach to Building Decision Trees Based on Goodman%2DKruskal Association Index.

Measures on Boolean polynomials and their applications in data mining.
 
Generalized Conditional Entropy and Decision Trees.

On Functions Defined on Free Boolean Algebras.
Abstract. We study the class of depth zero Boolean algebras, both from a classical viewpoint and an effective viewpoint. In particular, we provide an algebraic characterization, constructing an explicit measure for each depth zero Boolean algebra and demonstrating there are no others, and an effective characterization, providing a necessary and sufficient condition for a depth zero Boolean algebra of rank at most ω to have a computable presentation. 1.
Pruning Redundant Association Rules Using Maximum Entropy Principle.
Data mining algorithms produce huge sets of rules, practically impossible to analyze manually. It is thus important to develop methods for removing redundant rules from those sets. We present a solution to the problem using the Maximum Entropy approach. The problem of eciency of Maximum Entropy computations is addressed by using closed form solutions for the most frequent cases. Analytical and experimental evaluation of the proposed technique indicates that it eciently produces small sets of interesting association rules.
Support Approximations Using Bonferroni%2DType Inequalities.
The purpose of this paper is to examine the usability of Bonferroni-type combinatorial inequalities to estimation of support of itemsets as well as general Boolean expressions. Families of inequalities for various types of Boolean expressions are presented and evaluated experimentally.
An axiomatization of partition entropy.

An Axiomatization of Generalized Entropy of Partitions.
 
A General Measure of Rule Interestingness.
The paper presents a new general measure of rule interestingness. Many known measures such as chi&amp;sup2; , gini gain or entropy gain can be obtained from this measure by setting some numerical parameters representing the amount of trust we have in the estimates of certain probabilities from the data. Moreover we show that there is a continuum of measures having chi&amp;sup2; , Gini gain and entropy gain as boundary cases. Properties and experimental evaluation of the new measure are also presented.
Data Mining of Weak Functional Decompositions.

On Axiomatization of Conditional Entropy of Functions Between Finite Sets.

