Possibilistic estimation of distributions to leverage sparse data in machine learning

Andrea G.B. Tettamanzi, David Emsellem, Célia da Costa Pereira, Alessandro Venerandi, Giovanni Fusco

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

Prompted by an application in the area of human geography using machine learning to study housing market valuation based on the urban form, we propose a method based on possibility theory to deal with sparse data, which can be combined with any machine learning method to approach weakly supervised learning problems. More specifically, the solution we propose constructs a possibilistic loss function to account for an uncertain supervisory signal. Although the proposal is illustrated on a specific application, its basic principles are general. The proposed method is then empirically validated on real-world data.

Original languageEnglish
Title of host publicationInformation Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings
EditorsMarie-Jeanne Lesot, Susana Vieira, Marek Z. Reformat, João Paulo Carvalho, Anna Wilbik, Bernadette Bouchon-Meunier, Ronald R. Yager
Place of PublicationCham, Switzerland
PublisherSpringer
Pages431-444
Number of pages14
ISBN (Print)9783030501457
DOIs
Publication statusPublished - 19 Jun 2020
Event18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020 - Lisbon, Portugal
Duration: 15 Jun 202019 Jun 2020
https://ipmu2020.inesc-id.pt/

Publication series

NameCommunications in Computer and Information Science
Volume1237 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020
CountryPortugal
CityLisbon
Period15/06/2019/06/20
Internet address

Keywords

  • machine learning
  • possibility theory
  • weakly supervised learning

Fingerprint

Dive into the research topics of 'Possibilistic estimation of distributions to leverage sparse data in machine learning'. Together they form a unique fingerprint.

Cite this