Please use this identifier to cite or link to this item:
Files in This Item:
There are no files associated with this item.
Title: Aggregated outputs by linear models: An application on marine litter beaching prediction
Authors: Hernandez-Gonzalez, Jeronimo; Inza, Inaki; Granado, Igor and Basurko, Oihane C.; Fernandes, Jose A.; Lozano, Jose A.
Citation: INFORMATION SCIENCES, 2019, 481, 381-393
Abstract: In regression, a predictive model which is able to anticipate the output of a new case is learnt from a set of previous examples. The output or response value of these examples used for model training is known. When learning with aggregated outputs, the examples available for model training are individually unlabeled. Collectively, the aggregated outputs of different subsets of training examples are provided. In this paper, we propose an iterative methodology to learn linear models from this type of data. In spite of being simple, its competitive performance is shown in comparison with a straightforward solution and state-of-the-art techniques. A real world problem is also illustrated which naturally fits the aggregated outputs framework: the estimation of marine litter beaching along the southeast coastline of the Bay of Biscay. (C) 2019 Elsevier Inc. All rights reserved.
Keywords: Machine learning; Regression; Linear models; Aggregated outputs; Expectation-Maximization; Marine litter beaching; PLASTIC DEBRIS; INGESTION; ACCUMULATION; RIVERS
Issue Date: 2019
Type: Article
DOI: 10.1016/j.ins.2018.12.083
ISSN: 0020-0255
E-ISSN: 1872-6291
Funder: Basque GovernmentBasque Government [IT609-13]
Spanish Ministry of Economy and Competitiveness [TIN2016-78365-R]
BERC program 2018-2021 (Basque Government)Basque Government
Severo Ochoa Program (Spanish Ministry of Economy and Competitiveness) [SEV-2017-0718]
European Union (LIFE LEMA project)European Union (EU) [LIFE15/ENWES/000252]
Training of Technologists Program of the Department of Economic Development and Infrastructures of the Basque Government
Gipuzkoa Provincial Council
Appears in Publication types:Artículos científicos

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.