What question did this study set out to answer?

This research develops a new optimization model for multiclass classification that maximizes class separation while minimizing errors.

A unified optimization framework for multiclass classification with structured hyperplane arrangements

View Full Paper

Ask AI

Bookmark

Authors

Mathematical optimization enhances multiclass classification, suggesting improved efficiency and performance.

This research develops a new optimization model for multiclass classification that maximizes class separation while minimizing errors.
Proposed a mathematical optimization model for multiclass classification based on hyperplane arrangements.
Developed a kernel-based extension for constructing nonlinear decision boundaries.
Incorporated geometric structures like classification trees and discrete feature selection.
Devised a dynamic clustering matheuristic to handle large-scale instances.
Conducted computational experiments to assess performance against state-of-the-art methods.
Demonstrated computational efficiency compared to previous formulations.
Showed competitive classification performance on synthetic datasets and real-world benchmarks.
Validated the model's capacity to effectively incorporate various geometric structures.

Blanco et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

Mathematical optimization enhances multiclass classification, suggesting improved efficiency and performance.

This research develops a new optimization model for multiclass classification that maximizes class separation while minimizing errors.
Proposed a mathematical optimization model for multiclass classification based on hyperplane arrangements.
Developed a kernel-based extension for constructing nonlinear decision boundaries.
Incorporated geometric structures like classification trees and discrete feature selection.
Devised a dynamic clustering matheuristic to handle large-scale instances.
Conducted computational experiments to assess performance against state-of-the-art methods.
Demonstrated computational efficiency compared to previous formulations.
Showed competitive classification performance on synthetic datasets and real-world benchmarks.
Validated the model's capacity to effectively incorporate various geometric structures.

Blanco et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark