Special Issue "Classifier Ensembles: Efficient Techniques to Define Robust System Structures"
A special issue of Information (ISSN 2078-2489).
Deadline for manuscript submissions: 1 July 2019
One of the main challenges in the design of efficient classifier ensembles is the process of selecting their structure, especially with regard to individual classifiers with their respective parameters as well as combination methods. One way to improve the performance of classifier ensembles is by automatically selecting the best classifiers and combination method to compose their structure. This selection can be done statically or dynamically. In the static selection, once the ensemble structure is defined, all test instances will be classified by the same structure. In other words, the ensemble structure is selected before starting the classification step and used for all instances of this step. This selection is the most traditional and this problem has been treated as a meta-learning problem, automatic selection of machine learning (auto-ML) or an optimization problem. Recently, based on the assumption that every test instance has particularities and different levels of difficulties in the classification process, the dynamic selection approach has arisen. In this selection, each instance is classified by a different ensemble structure (set of individual classifiers and method of combination). Thus, in order to classify N instances it is necessary to set N ensemble structures, one for each test instance, selecting the most appropriate structure to classify one specific instance. In dynamic selection, the selection of the ensemble structure is performed during the testing phase.
Despite the high number of studies and techniques, finding an optimal parameter set that maximizes classification accuracy of an ensemble system is still an open problem. The search space for all parameters of an ensemble system (classifier type, size, classifier parameters, combination method and feature selection) is very large. This Special Issue on “Classifier Ensembles: Efficient Techniques to Define Robust System Structures” aims to promote new theories, techniques, and methods with which to exploit the selection of efficient ensemble structures, including both selection approaches, static and dynamic selections.
Dr. Anne Canuto
Manuscript Submission Information
Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.
Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly journal published by MDPI.
Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.
- Classifier Ensembles
- Optimization Techniques
- Dynamic Selection in Classifier Ensembles