The Effect of Imbalanced Data Class Distribution on Fuzzy Classifiers - Experimental Study
Published 27 July 2005
Sofia Visa, Anca Ralescu
Citations33
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
The experimental results reported here show that fuzzy classifiers are less variant with the class distribution and less sensitive to the imbalance factor than decision trees.
Abstract
This study evaluates the robustness of a fuzzy classifier when class distribution of the training set varies. The analysis of the results is based on the classification accuracy and ROC curves. The experimental results reported here show that fuzzy classifiers are less variant with the class distribution and less sensitive to the imbalance factor than decision trees
Keywords
Computer Science
The Case against Accuracy Estimation for Comparing Induction Algorithms
1,024 Citations1998Foster Provost, Tom Fawcett +1 more
This work describes and demonstrates what it believes to be the proper use of ROC analysis for comparative studies in machine learning research, and argues that this methodology is preferable both for making practical choices and for drawing conclusions.
Research Studies Press eBooksFril- Fuzzy and Evidential Reasoning in Artificial Intelligence
240 Citations1995J.F. Baldwin, T. P. Martin +1 more
This book extends the logic programming form of knowledge representation and method of inference to permit the inclusion of uncertainties such as probabilistic knowledge and fuzzy incompleteness to general areas of knowledge engineering including expert and decision-support systems, evidential and case-based reasoning, fuzzy control and databases.
Fuzzy Sets and SystemsExtensions of fuzzy aggregation
53 Citations1997Anca Ralescu, Dan A. Ralescu
It is argued that, for the purpose of aggregating fuzzy sets, comonotonic maxitivity is a more desirable requirement than comonOTonic additivity, and in the absence of any such requirement, a wider class of aggregation procedures are explored.
The effect of small disjuncts and class distribution on decision tree learning
41 Citations2003Gary M. Weiss, Haym Hirsh
The experimental results indicate that the naturally occurring class distribution is not always best for learning and that a balanced class distribution should be chosen to generate a classifier robust to different misclassification costs.
Elsevier eBooksLinguistic Modeling of Physical Task Characteristics
2 Citations2003Sofia Visa, Anca Ralescu +2 more
This chapter describes the approach for linguistic modeling based on fuzzy sets applied to modeling the meaning of linguistic descriptions for lifting task variables and addresses the problem of modeling human perception of the linguistic descriptions of numerical values.