Theoretical Study of the Relationship between Diversity and Single-Class Measures for Class Imbalance Learning
Published 1 December 2009
Shuo Wang, Xin Yao
Citations12
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
One of the main objectives of this paper is to find out if and when ensemble diversity can improve the classification performance on the important (minority) class.
Abstract
This paper presents the theoretical research about the relationship between diversity of classification ensembles and single-class measures that are commonly used in class imbalance learning. Although there have been studies on diversity and its links to overall ensemble accuracy, little work has been done on the impact of diversity on single-class performance measures in class imbalance learning. The study of class imbalance learning is important, because many real-world problems, such as those in medical diagnosis, fraud detection, condition monitoring, etc., have imbalanced classes, where a minority class is usually more important and interesting than the majority class. In order to gain a deeper understanding of ensemble learning for imbalanced classes, this paper studies the impact of diversity on single-class performance measures theoretically and empirically. One of the main objectives of this paper is to find out if and when ensemble diversity can improve the classification performance on the important (minority) class.
Keywords
Computer ScienceEngineeringBusiness, Management and Accounting
Lecture notes in computer scienceSMOTEBoost: Improving Prediction of the Minority Class in Boosting
1,633 Citations2003Nitesh V. Chawla, Aleksandar Lazarević +2 more
This paper presents a novel approach for learning from imbalanced data sets, based on a combination of the SMOTE algorithm and the boosting procedure, which shows improvement in prediction performance on the minority class and overall improved F-values.
Handling imbalanced datasets: A review
880 Citations2006Sotiris Kotsiantis, Dimitris Kanellopoulos +1 more
This paper describes various techniques for handling imbalanced dataset problems, and hopes that the references cited will cover the major theoretical issues, guiding the researcher in interesting re- search directions and suggesting possible bias combinations that have yet to be explored.
ACM SIGKDD Explorations NewsletterLearning from imbalanced data sets with boosting and data generation
606 Citations2004Hongyu Guo, Herna L. Viktor
The approach does not sacrifice one class in favor of the other, but produces high predictions against both minority and majority classes, and compares well in comparison with a base classifier, a standard benchmarking boosting algorithm and three advanced boosting-based algorithms for imbalanced data set.
AdaCost: Misclassification Cost-Sensitive Boosting
596 Citations1999Wei Fan, Salvatore J. Stolfo +2 more
It is formally show that AdaCost reduces the upper bound of cumulative misclassification cost of the training set of the training set without consuming additional computing power.
Diversity analysis on imbalanced data sets by using ensemble models
546 Citations2009Shuo Wang, Xin Yao
This paper explores the impact of diversity on each class and overall performance and improves SMOTE in a novel way for solving multi-class data sets in ensemble model - SMOTEBagging.
Philosophical Transactions of the Royal Society of London Series A Containing Papers of a Mathematical or Physical CharacterVII. On the association of attributes in statistics: with illustrations from the material of the childhood society, &c
400 Citations1900George Yule
In the ordinary theory of statistical correlation, normal or otherwise, material susceptible of continuous variation, or at least of variation by a considerable number of discontinuous steps, certain practical cases arise, where either no variation is thinkable at all, or else is not measured or possibly measurable.
Pattern Analysis and ApplicationsLimits on the majority vote accuracy in classifier fusion
352 Citations2003Ludmila I. Kuncheva, Christopher J. Whitaker +2 more
It is shown that the majority vote with dependent classifiers can potentially offer a dramatic improvement both over independent classifiers and over an individual classifier with accuracy p, and a functional relationship between the limits and the pairwise dependence Q is derived.
Journal of Machine Learning ResearchManaging Diversity in Regression Ensembles
345 Citations2005Gavin Brown, Jeremy Wyatt +1 more
It is demonstrated that these methods control the bias-variance-covariance trade-off systematically, and can be utilised with any estimator capable of minimising a quadratic error function, for example MLPs, or RBF networks.
Proceedings of International Conference on Neural Networks (ICNN'96)Generalization error of ensemble estimators
236 Citations2002N. Ueda, Ryoko Nakano
An analytical result is presented for the generalization error of ensemble estimators by using factors of interest (bias, variance, covariance, and noise variance) and showing how thegeneralization error is affected by each of them.
Ten measures of diversity in classifier ensembles: limits for two classifiers
140 Citations2001Ludmila I. Kuncheva
10 measures of classifier diversity: 4 pairwise and 6 non-pairwise measures and the limits of the measures for 2 classifiers of equal accuracy are derived.
arXiv (Cornell University)Linear and Order Statistics Combiners for Pattern Classification
137 Citations1999Kagan Tumer, Joydeep Ghosh
Lecture notes in computer scienceThat Elusive Diversity in Classifier Ensembles
127 Citations2003Ludmila I. Kuncheva
It is argued that even without a clear-cut definition and theory behind it, studying diversity may prompt viable heuristic solutions in classifier ensembles.
ACM SIGIR ForumInformation retrieval
125 Citations1983C. J. van Rijsbergen
It is asked whether this difference between 'aboutness' and 'topicality' should be taken into account when designing IR systems, and if so, how to take it into account.
Classifying imbalanced data using a bagging ensemble variation (BEV)
85 Citations2007Cen Li
The motivation behind the scheme is to maximally use the minority class data without creating synthetic data or making changes to the existing classification systems, and experimental results show the effectiveness of the system.
Perspectives in neural computingLinear and Order Statistics Combiners for Pattern Classification
56 Citations1999Amanda Sharkey
This chapter provides an analytical framework to quantify the improvements in classification results due to combining and derives expressions that indicate how much the median, the maximum and in general the i-th order statistic can improve classifier performance.
Lecture notes in computer scienceExploiting Diversity in Ensembles: Improving the Performance on Unbalanced Datasets
51 Citations2007Nitesh V. Chawla, Jared Sylvester
A genetic algorithm based framework is implemented to weight the contribution of each classifier by an appropriate fitness function, such that the classifiers that complement each other on the unbalanced dataset are preferred, resulting in significantly improved performances.
Diversity exploration and negative correlation learning on imbalanced data sets
32 Citations2009Shuo Wang, Ke Tang +1 more
This paper investigates if diversity can relieve the class imbalance problem by using negative correlation learning (NCL) model, which encourages diversity explicitly by adding a penalty term in the error function of neural networks.