Search research articles

Search research articles

Related Experiment Videos

An optimization model for constrained discriminant analysis and numerical experiments with iris, thyroid, and heart

R J Gallagher¹, E K Lee, D A Patterson

¹Department of Medical Informatics, Columbia University, New York, New York, USA.

Proceedings : a Conference of the American Medical Informatics Association. AMIA Fall Symposium

|January 1, 1996

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Limits on WIMP Dark Matter with NaI(Tl) Crystals in Three Years of COSINE-100 Data.

Physical review letters·2025

Same author

Combined Annual Modulation Dark Matter Search with COSINE-100 and ANAIS-112.

Physical review letters·2025

Same author

Improved Limit on Neutrinoless Double Beta Decay of ^{100}Mo from AMoRE-I.

Physical review letters·2025

Same author

Using vitality indicators to predict survival of aquatic animals released from fisheries.

Conservation physiology·2024

Same author

Search for Boosted Dark Matter in COSINE-100.

Physical review letters·2023

Same author

Detection efficiency calibration for an array of fourteen HPGe detectors.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2023

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

A new nonlinear programming model controls misclassification probabilities in discriminant analysis. This method allows for restricted errors and a reserved-judgment region for uncertain classifications.

Area of Science:

Operations Research
Statistical Classification
Machine Learning

Background:

Discriminant analysis is crucial for classifying data points into predefined groups.
Existing methods may lack precise control over misclassification rates.
Handling uncertain classifications requires robust algorithmic approaches.

Purpose of the Study:

To present a nonlinear 0/1 mixed integer programming model for constrained discriminant analysis.
To enable explicit control over misclassification probabilities within the model.
To introduce a mechanism for allocating uncertain data points to a reserved-judgment region.

Main Methods:

Formulation of a nonlinear 0/1 mixed integer programming model.
Incorporation of constraints on the number of allowed misclassifications.

Related Experiment Videos

Development of a linearization technique for the proposed model.

Application and testing on medical and non-medical datasets.

Main Results:

The model successfully controls misclassification probabilities through defined restrictions.
A "reserved-judgment" region effectively handles entities with ambiguous classifications.
Preliminary numerical results demonstrate the model's applicability in diverse domains.

Conclusions:

The developed nonlinear programming model offers a powerful tool for constrained discriminant analysis.
The approach provides enhanced control over classification accuracy and uncertainty management.
The model shows promise for applications in fields requiring precise data classification, such as medicine.