Welcome to the InfoSci Platform
Reference Hub1
Learning Verifiable Ensembles for Classification Problems with High Safety Requirements

Learning Verifiable Ensembles for Classification Problems with High Safety Requirements

Sebastian Nusser, Clemens Otte, Werner Hauptmann, Rudolf Kruse
Copyright: © 2010 |Pages: 27
ISBN13: 9781615207572|ISBN10: 1615207570|EISBN13: 9781615207589
DOI: 10.4018/978-1-61520-757-2.ch019
Cite Chapter Cite Chapter

MLA

Nusser, Sebastian, et al. "Learning Verifiable Ensembles for Classification Problems with High Safety Requirements." Intelligent Soft Computation and Evolving Data Mining: Integrating Advanced Technologies, edited by Leon Shyue-Liang Wang and Tzung-Pei Hong, IGI Global, 2010, pp. 405-431. https://doi.org/10.4018/978-1-61520-757-2.ch019

APA

Nusser, S., Otte, C., Hauptmann, W., & Kruse, R. (2010). Learning Verifiable Ensembles for Classification Problems with High Safety Requirements. In L. Wang & T. Hong (Eds.), Intelligent Soft Computation and Evolving Data Mining: Integrating Advanced Technologies (pp. 405-431). IGI Global. https://doi.org/10.4018/978-1-61520-757-2.ch019

Chicago

Nusser, Sebastian, et al. "Learning Verifiable Ensembles for Classification Problems with High Safety Requirements." In Intelligent Soft Computation and Evolving Data Mining: Integrating Advanced Technologies, edited by Leon Shyue-Liang Wang and Tzung-Pei Hong, 405-431. Hershey, PA: IGI Global, 2010. https://doi.org/10.4018/978-1-61520-757-2.ch019

Export Reference

Mendeley
Favorite

Abstract

This chapter describes a machine learning approach for classification problems in safety-related domains. The proposed method is based on ensembles of low-dimensional submodels. The usage of low-dimensional submodels enables the domain experts to understand the mechanisms of the learned solution. Due to the limited dimensionality of the submodels each individual model can be visualized and can thus be interpreted and validated according to the domain knowledge. The ensemble of all submodels overcomes the limited predictive performance of each single submodel while the overall solution remains interpretable and verifiable. By different examples from real-world applications the authors will show that their classification approach is applicable to a wide range of classification problems in the field of safety-related applications - ranging from decision support systems over plant monitoring and diagnosis systems to control tasks with very high safety requirements.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.