Cargando…

Cautious Label-Wise Ranking with Constraint Satisfaction

Ranking problems are difficult to solve due to their combinatorial nature. One way to solve this issue is to adopt a decomposition scheme, splitting the initial difficult problem in many simpler problems. The predictions obtained from these simplified settings must then be combined into one single o...

Descripción completa

Detalles Bibliográficos
Autores principales: Carranza-Alarcon, Yonatan-Carlos, Messoudi, Soundouss, Destercke, Sébastien
Formato: Online Artículo Texto
Lenguaje:English
Publicado: 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7274705/
http://dx.doi.org/10.1007/978-3-030-50143-3_8
_version_ 1783542642667159552
author Carranza-Alarcon, Yonatan-Carlos
Messoudi, Soundouss
Destercke, Sébastien
author_facet Carranza-Alarcon, Yonatan-Carlos
Messoudi, Soundouss
Destercke, Sébastien
author_sort Carranza-Alarcon, Yonatan-Carlos
collection PubMed
description Ranking problems are difficult to solve due to their combinatorial nature. One way to solve this issue is to adopt a decomposition scheme, splitting the initial difficult problem in many simpler problems. The predictions obtained from these simplified settings must then be combined into one single output, possibly resolving inconsistencies between the outputs. In this paper, we consider such an approach for the label ranking problem, where in addition we allow the predictive model to produce cautious inferences in the form of sets of rankings when it lacks information to produce reliable, precise predictions. More specifically, we propose to combine a rank-wise decomposition, in which every sub-problem becomes an ordinal classification one, with a constraint satisfaction problem (CSP) approach to verify the consistency of the predictions. Our experimental results indicate that our approach produces predictions with appropriately balanced reliability and precision, while remaining competitive with classical, precise approaches.
format Online
Article
Text
id pubmed-7274705
institution National Center for Biotechnology Information
language English
publishDate 2020
record_format MEDLINE/PubMed
spelling pubmed-72747052020-06-08 Cautious Label-Wise Ranking with Constraint Satisfaction Carranza-Alarcon, Yonatan-Carlos Messoudi, Soundouss Destercke, Sébastien Information Processing and Management of Uncertainty in Knowledge-Based Systems Article Ranking problems are difficult to solve due to their combinatorial nature. One way to solve this issue is to adopt a decomposition scheme, splitting the initial difficult problem in many simpler problems. The predictions obtained from these simplified settings must then be combined into one single output, possibly resolving inconsistencies between the outputs. In this paper, we consider such an approach for the label ranking problem, where in addition we allow the predictive model to produce cautious inferences in the form of sets of rankings when it lacks information to produce reliable, precise predictions. More specifically, we propose to combine a rank-wise decomposition, in which every sub-problem becomes an ordinal classification one, with a constraint satisfaction problem (CSP) approach to verify the consistency of the predictions. Our experimental results indicate that our approach produces predictions with appropriately balanced reliability and precision, while remaining competitive with classical, precise approaches. 2020-05-15 /pmc/articles/PMC7274705/ http://dx.doi.org/10.1007/978-3-030-50143-3_8 Text en © Springer Nature Switzerland AG 2020 This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
spellingShingle Article
Carranza-Alarcon, Yonatan-Carlos
Messoudi, Soundouss
Destercke, Sébastien
Cautious Label-Wise Ranking with Constraint Satisfaction
title Cautious Label-Wise Ranking with Constraint Satisfaction
title_full Cautious Label-Wise Ranking with Constraint Satisfaction
title_fullStr Cautious Label-Wise Ranking with Constraint Satisfaction
title_full_unstemmed Cautious Label-Wise Ranking with Constraint Satisfaction
title_short Cautious Label-Wise Ranking with Constraint Satisfaction
title_sort cautious label-wise ranking with constraint satisfaction
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7274705/
http://dx.doi.org/10.1007/978-3-030-50143-3_8
work_keys_str_mv AT carranzaalarconyonatancarlos cautiouslabelwiserankingwithconstraintsatisfaction
AT messoudisoundouss cautiouslabelwiserankingwithconstraintsatisfaction
AT desterckesebastien cautiouslabelwiserankingwithconstraintsatisfaction