Approaches for Multi-View Redescription Mining

Abstract

The task of redescription mining explores ways to re-describe different subsets of entities contained in a dataset and to reveal non-trivial associations between different subsets of attributes, called views. This interesting and challenging task is encountered in different scientific fields, and is addressed by a number of approaches that obtain redescriptions and allow for the exploration and analyses of attribute associations. The main limitation of existing approaches to this task is their inability to use more than two views. Our work alleviates this drawback. We present a memory efficient, extensible multi-view redescription mining framework that can be used to relate multiple, i.e. more than two views, disjoint sets of attributes describing one set of entities. The framework can use any multi-target regression or multi-label classification algorithm, with models that can be represented as sets of rules, to generate redescriptions. Multi-view redescriptions are built using incremental view-extending heuristic from initially created twoview redescriptions. In this work, we use different types of Predictive Clustering trees algorithms (regular, extra, with random output selection) and the Random Forest thereof in order to improve the quality of final redescription sets and/or execution time needed to generate them. We provide multiple performance analyses of the proposed framework and compare it against the naive approach to multi-view redescription mining. We demonstrate the usefulness of the proposed multi-view extension on several datasets, including a use-case on understanding of machine learning models-a topic of growing importance in machine learning and artificial intelligence in general. INDEX TERMS knowledge discovery, multi-view redescription mining, redescription set, predictive clustering trees, random forest, extremely randomized trees, random output selection.

Heavy Metals and Essential Metals Are Associated with Cerebrospinal Fluid Biomarkers of Alzheimer’s Disease

“…In this work, we use the minimal accuracy threshold of 0.5. The statistical significance of a redescription (reported through a corresponding p-value) measures how probable it would be to obtain a redescription at random (by a random choice of rules that form it), so that each rule in a randomly created redescription describes the same number of patients as the original, and that the resulting redescription has a support set size equal or larger to the support set size of the original redescription [93]. In this work, we use a maximal p-value of 0.01.…”