A tutorial on statistically sound pattern discovery

DOI10.1007/S10618-018-0590-XzbMATH Open1464.62305arXiv1709.03904OpenAlexW3103529187WikidataQ128730535 ScholiaQ128730535MaRDI QIDQ2218330FDOQ2218330

Authors: Wilhelmiina Hämäläinen, Geoffrey I. Webb 

Publication date: 15 January 2021

Published in: Data Mining and Knowledge Discovery (Search for Journal in Brave)

Abstract: Statistically sound pattern discovery harnesses the rigour of statistical hypothesis testing to overcome many of the issues that have hampered standard data mining approaches to pattern discovery. Most importantly, application of appropriate statistical tests allows precise control over the risk of false discoveries -- patterns that are found in the sample data but do not hold in the wider population from which the sample was drawn. Statistical tests can also be applied to filter out patterns that are unlikely to be useful, removing uninformative variations of the key patterns in the data. This tutorial introduces the key statistical and data mining theory and techniques that underpin this fast developing field. We concentrate on two general classes of patterns: dependency rules that express statistical dependencies between condition and consequent parts and dependency sets that express mutual dependence between set elements. We clarify alternative interpretations of statistical dependence and introduce appropriate tests for evaluating statistical significance of patterns in different situations. We also introduce special techniques for controlling the likelihood of spurious discoveries when multitudes of patterns are evaluated. The paper is aimed at a wide variety of audiences. It provides the necessary statistical background and summary of the state-of-the-art for any data mining researcher or practitioner wishing to enter or understand statistically sound pattern discovery research or practice. It can serve as a general introduction to the field of statistically sound pattern discovery for any reader with a general background in data sciences.

Full work available at URL: https://arxiv.org/abs/1709.03904

Recommendations

• Discovering significant patterns
• A statistical significance testing approach to mining the most informative set of patterns
• Pattern Discovery and Detection: A Unified Statistical Methodology
• Significance tests for unsupervised pattern discovery in large continuous multivariate data sets
• Layered critical values: a powerful direct-adjustment approach to discovering significant patterns

zbMATH Keywords

hypothesis testingassociation rulestatistical significancepattern discoverydependency ruledependency set

Mathematics Subject Classification ID

Statistical aspects of big data and data science (62R07) Hypothesis testing in multivariate analysis (62H15) Pattern recognition, speech recognition (68T10) Image analysis in multivariate analysis (62H35)

Cites Work

• Title not available (Why is that?)
• Test of Significance for 2 × 2 Contingency Tables
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Testing Statistical Hypotheses
• Title not available (Why is that?)
• THE MEANING OF A SIGNIFICANCE LEVEL
• Resampling-based multiple testing for microarray data analysis (With comments)
• The control of the false discovery rate in multiple testing under dependency.
• A sharper Bonferroni procedure for multiple tests of significance
• Multiple Hypotheses Testing with Weights
• Asymptotic optimality of the Westfall-Young permutation procedure for multiple testing under dependence
• Title not available (Why is that?)
• Rectangular Confidence Regions for the Means of Multivariate Normal Distributions
• Genome-wide significance levels and weighted hypothesis testing
• A survey of exact inference for contingency tables. With comments and a rejoinder by the author
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Statistics
• Supervised descriptive rule discovery: a unifying survey of contrast set, emerging pattern and subgroup mining
• Title not available (Why is that?)
• The Fisher, Neyman-Pearson Theories of Testing Hypotheses: One Theory or Two?
• An Application of Markov Chain Monte Carlo to Community Ecology
• Redundancy, deduction schemes, and minimum-size bases for association rules
• Bayesian inference for categorical data analysis
• Discovering significant patterns
• Frequent Pattern Mining
• Bayesian Testing and Estimation of Association in a Two-Way Contingency Table
• Knowledge Discovery in Inductive Databases
• Bayesian statistics. An introduction.
• Detecting group differences: Mining contrast sets
• Multiple testing for exploratory research
• Interesting patterns
• On permutation procedures for strong control in multiple testing with gene expression data
• Frequentist Performance of Bayesian Confidence Intervals for Comparing Proportions in 2 × 2 Contingency Tables
• Layered critical values: a powerful direct-adjustment approach to discovering significant patterns
• Title not available (Why is that?)
• A Comparison of Some Continuity Corrections for the Chi-Squared Test on 2 × 2 Tables
• The \(2\times2\) table: A discussion from a Bayesian viewpoint.
• New upper bounds for tight and fast approximation of Fisher's exact test in dependency rule mining
• Efficient search methods for statistical dependency rules
• FDR- and FWE-controlling methods using data-driven weights
• HOW GOOD IS A NORMAL APPROXIMATION FOR RATES AND PROPORTIONS OF LOW INCIDENCE EVENTS?
• Uses, Abuses and Misuses of Significance Tests in the Scientific Community: Won't the Bayesian Choice be Unavoidable?
• Statistical significance of combinatorial regulations

Cited In (12)

• Significance tests for unsupervised pattern discovery in large continuous multivariate data sets
• Robust subgroup discovery. Discovering subgroup lists using MDL
• The minimum description length principle for pattern mining: a survey
• Statistical inference and data mining: false discoveries control
• Pattern Discovery and Detection: A Unified Statistical Methodology
• Layered critical values: a powerful direct-adjustment approach to discovering significant patterns
• A statistical significance testing approach to mining the most informative set of patterns
• SPEck: mining statistically-significant sequential patterns efficiently with exact sampling
• Efficient search methods for statistical dependency rules
• Title not available (Why is that?)
• ROhAN: row-order agnostic null models for statistically-sound knowledge discovery
• Discovering significant patterns

Uses Software

• UCI-ml
• sedaR
• CMAR
• TEIRESIAS

This page was built for publication: A tutorial on statistically sound pattern discovery

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2218330)