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

doi:10.1007/978-3-030-50146-4_32

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

Architecture

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

2 Downloads (Pure)

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 language	English
Title of host publication	Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings
Editors	Marie-Jeanne Lesot, Susana Vieira, Marek Z. Reformat, João Paulo Carvalho, Anna Wilbik, Bernadette Bouchon-Meunier, Ronald R. Yager
Place of Publication	Cham, Switzerland
Publisher	Springer
Pages	431-444
Number of pages	14
ISBN (Print)	9783030501457
DOIs	https://doi.org/10.1007/978-3-030-50146-4_32
Publication status	Published - 19 Jun 2020
Event	18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020 - Lisbon, Portugal Duration: 15 Jun 2020 → 19 Jun 2020 https://ipmu2020.inesc-id.pt/

Publication series

Name	Communications in Computer and Information Science
Volume	1237 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020
Country/Territory	Portugal
City	Lisbon
Period	15/06/20 → 19/06/20
Internet address	https://ipmu2020.inesc-id.pt/

Keywords

machine learning
possibility theory
weakly supervised learning

Access to Document

10.1007/978-3-030-50146-4_32

Tettamanzi-etal-IPMU-2020-Possibilistic-estimation-of-distributions-to-leverage-sparse-data-in-machine-learningAccepted author manuscript, 666 KB

Cite this

Tettamanzi, A. G. B., Emsellem, D., da Costa Pereira, C., Venerandi, A., & Fusco, G. (2020). Possibilistic estimation of distributions to leverage sparse data in machine learning. In M-J. Lesot, S. Vieira, M. Z. Reformat, J. P. Carvalho, A. Wilbik, B. Bouchon-Meunier, & R. R. Yager (Eds.), Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings (pp. 431-444). (Communications in Computer and Information Science; Vol. 1237 CCIS). Springer. https://doi.org/10.1007/978-3-030-50146-4_32

Tettamanzi, Andrea G.B. ; Emsellem, David ; da Costa Pereira, Célia et al. / Possibilistic estimation of distributions to leverage sparse data in machine learning. Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings. editor / Marie-Jeanne Lesot ; Susana Vieira ; Marek Z. Reformat ; João Paulo Carvalho ; Anna Wilbik ; Bernadette Bouchon-Meunier ; Ronald R. Yager. Cham, Switzerland : Springer, 2020. pp. 431-444 (Communications in Computer and Information Science).

@inproceedings{b5b021c004ed4409ab2299612f94aae3,

title = "Possibilistic estimation of distributions to leverage sparse data in machine learning",

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.",

keywords = "machine learning, possibility theory, weakly supervised learning",

author = "Tettamanzi, {Andrea G.B.} and David Emsellem and {da Costa Pereira}, C{\'e}lia and Alessandro Venerandi and Giovanni Fusco",

year = "2020",

month = jun,

day = "19",

doi = "10.1007/978-3-030-50146-4_32",

language = "English",

isbn = "9783030501457",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "431--444",

editor = "Marie-Jeanne Lesot and Susana Vieira and Reformat, {Marek Z.} and Carvalho, {Jo{\~a}o Paulo} and Anna Wilbik and Bernadette Bouchon-Meunier and Yager, {Ronald R.}",

booktitle = "Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings",

note = "18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020 ; Conference date: 15-06-2020 Through 19-06-2020",

url = "https://ipmu2020.inesc-id.pt/",

}

Tettamanzi, AGB, Emsellem, D, da Costa Pereira, C, Venerandi, A & Fusco, G 2020, Possibilistic estimation of distributions to leverage sparse data in machine learning. in M-J Lesot, S Vieira, MZ Reformat, JP Carvalho, A Wilbik, B Bouchon-Meunier & RR Yager (eds), Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings. Communications in Computer and Information Science, vol. 1237 CCIS, Springer, Cham, Switzerland, pp. 431-444, 18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020, Lisbon, Portugal, 15/06/20. https://doi.org/10.1007/978-3-030-50146-4_32

Possibilistic estimation of distributions to leverage sparse data in machine learning. / Tettamanzi, Andrea G.B.; Emsellem, David; da Costa Pereira, Célia et al.

Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings. ed. / Marie-Jeanne Lesot; Susana Vieira; Marek Z. Reformat; João Paulo Carvalho; Anna Wilbik; Bernadette Bouchon-Meunier; Ronald R. Yager. Cham, Switzerland : Springer, 2020. p. 431-444 (Communications in Computer and Information Science; Vol. 1237 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution book

TY - GEN

T1 - Possibilistic estimation of distributions to leverage sparse data in machine learning

AU - Tettamanzi, Andrea G.B.

AU - Emsellem, David

AU - da Costa Pereira, Célia

AU - Venerandi, Alessandro

AU - Fusco, Giovanni

PY - 2020/6/19

Y1 - 2020/6/19

N2 - 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.

AB - 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.

KW - machine learning

KW - possibility theory

KW - weakly supervised learning

U2 - 10.1007/978-3-030-50146-4_32

DO - 10.1007/978-3-030-50146-4_32

M3 - Conference contribution book

AN - SCOPUS:85086235064

SN - 9783030501457

T3 - Communications in Computer and Information Science

SP - 431

EP - 444

BT - Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings

A2 - Lesot, Marie-Jeanne

A2 - Vieira, Susana

A2 - Reformat, Marek Z.

A2 - Carvalho, João Paulo

A2 - Wilbik, Anna

A2 - Bouchon-Meunier, Bernadette

A2 - Yager, Ronald R.

PB - Springer

CY - Cham, Switzerland

T2 - 18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2020

Y2 - 15 June 2020 through 19 June 2020

ER -

Tettamanzi AGB, Emsellem D, da Costa Pereira C, Venerandi A, Fusco G. Possibilistic estimation of distributions to leverage sparse data in machine learning. In Lesot M-J, Vieira S, Reformat MZ, Carvalho JP, Wilbik A, Bouchon-Meunier B, Yager RR, editors, Information Processing and Management of Uncertainty in Knowledge-Based Systems - 18th International Conference, IPMU 2020, Proceedings. Cham, Switzerland: Springer. 2020. p. 431-444. (Communications in Computer and Information Science). Epub 2020 Jun 5. doi: 10.1007/978-3-030-50146-4_32

Possibilistic estimation of distributions to leverage sparse data in machine learning

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this