Reducing Misclassification Costs

Pazzani, Michael J.; Merz, Christopher J.; Murphy, Patrick M.; Ali, Kamal; Hume, Timothy; Brunk, Clifford

doi:10.1016/b978-1-55860-335-6.50034-9

Cited by 258 publications

(173 citation statements)

References 8 publications

Supporting

Mentioning

165

Contrasting

Unclassified

Order By: Relevance

“…This can be seen as a special case of multitask learning (Caruana, 1993), the more general idea being to define supplementary classification tasks in which the classes are more equally balanced. Pazzani et al (1994) assign different weights to examples of different classes, Fawcett and Provost (1997) prune the possibly overfit rule set learned from an imbalanced set, and Ezawa et al (1996) force the learner to consider relationships between certain attributes above others.…”

Section: Methodological Issuesmentioning

confidence: 99%

“…The standard decision theoretic approach to defining the "optimal" tradeoff between false and true positives is to assign relative costs to errors of omission and errors of commission, and to make the classification that minimizes expected cost (Pazzani et al, 1994). One deterrent to using this approach is that the costs are often hard to determine and may involve multiple considerations whose units are incommensurable (e.g., monetary cost, pollution levels, international reputation).…”

Section: Performance Measurementioning

confidence: 99%

See 1 more Smart Citation

Untitled

1998

View full text Add to dashboard Cite

Abstract. During a project examining the use of machine learning techniques for oil spill detection, we encountered several essential questions that we believe deserve the attention of the research community. We use our particular case study to illustrate such issues as problem formulation, selection of evaluation measures, and data preparation. We relate these issues to properties of the oil spill application, such as its imbalanced class distribution, that are shown to be common to many applications. Our solutions to these issues are implemented in the Canadian Environmental Hazards Detection System (CEHDS), which is about to undergo field testing.

show abstract

Section: Methodological Issuesmentioning

confidence: 99%

Section: Performance Measurementioning

confidence: 99%

Untitled

1998

View full text Add to dashboard Cite

show abstract

“…One is to re-sample an original dataset [11], [14], [15], either by oversampling a minority class or under-sampling a majority class until two classes are nearly balanced. The second is to use cost sensitive learning by assigning distinct costs to correctly classified instances or classifications errors [7], [9], [16]. (1)…”

Section: Introductionmentioning

confidence: 99%

Safe-Level-SMOTE: Safe-Level-Synthetic Minority Over-Sampling TEchnique for Handling the Class Imbalanced Problem

Bunkhumpornpat

Sinapiromsaran

Lursinsap

2009

Lecture Notes in Computer Science

724

317

View full text Add to dashboard Cite

Abstract. The class imbalanced problem occurs in various disciplines when one of target classes has a tiny number of instances comparing to other classes. A typical classifier normally ignores or neglects to detect a minority class due to the small number of class instances. SMOTE is one of over-sampling techniques that remedies this situation. It generates minority instances within the overlapping regions. However, SMOTE randomly synthesizes the minority instances along a line joining a minority instance and its selected nearest neighbours, ignoring nearby majority instances. Our technique called SafeLevel-SMOTE carefully samples minority instances along the same line with different weight degree, called safe level. The safe level computes by using nearest neighbour minority instances. By synthesizing the minority instances more around larger safe level, we achieve a better accuracy performance than SMOTE and Borderline-SMOTE.

show abstract

“…Unfortunately, little work has been published on either problem. There exist several dozen articles in which techniques for costsensitive learning are suggested (Turney, 1996), but few studies evaluate and compare them (Domingos, 1999;Pazzani et al, 1994;Provost, Fawcett, & Kohavi, 1998). The literature provides even less guidance in situations where distributions are imprecise or can change.…”

Section: Classifier Comparison: Decision Analysis and Roc Analysismentioning

confidence: 99%

Untitled

2001

View full text Add to dashboard Cite

Abstract. In real-world environments it usually is difficult to specify target operating conditions precisely, for example, target misclassification costs. This uncertainty makes building robust classification systems problematic. We show that it is possible to build a hybrid classifier that will perform at least as well as the best available classifier for any target conditions. In some cases, the performance of the hybrid actually can surpass that of the best known classifier. This robust performance extends across a wide variety of comparison frameworks, including the optimization of metrics such as accuracy, expected cost, lift, precision, recall, and workforce utilization. The hybrid also is efficient to build, to store, and to update. The hybrid is based on a method for the comparison of classifier performance that is robust to imprecise class distributions and misclassification costs. The ROC convex hull (ROCCH) method combines techniques from ROC analysis, decision analysis and computational geometry, and adapts them to the particulars of analyzing learned classifiers. The method is efficient and incremental, minimizes the management of classifier performance data, and allows for clear visual comparisons and sensitivity analyses. Finally, we point to empirical evidence that a robust hybrid classifier indeed is needed for many real-world problems.

show abstract

Reducing Misclassification Costs

Cited by 258 publications

References 8 publications

Untitled

Untitled

Safe-Level-SMOTE: Safe-Level-Synthetic Minority Over-Sampling TEchnique for Handling the Class Imbalanced Problem

Untitled

Contact Info

Product

Resources

About