Designing multi-label classifiers that maximize F measures: state of the art

PILLAI, IGNAZIO;FUMERA, GIORGIO;ROLI, FABIO
2017

Abstract

Multi-label classification problems usually occur in tasks related to information retrieval, like text and image annotation, and are receiving increasing attention from the machine learning and pattern recognition fields. One of the main issues under investigation is the development of classification algorithms capable of maximizing specific accuracy measures based on precision and recall. We focus on the widely used F measure, defined for binary, single-label problems as the weighted harmonic mean of precision and recall, and later extended to multi-label problems in three ways: macro-averaged, micro-averaged and instance-wise. In this paper we give a comprehensive survey of theoretical results and algorithms aimed at maximizing F measures. We subdivide it according to the two main existing approaches: empirical utility maximization, and decision-theoretic. Under the former approach, we also derive the optimal (Bayes) classifier at the population level for the instance-wise and micro-averaged F, extending recent results about the single-label F. In a companion paper we shall focus on the micro-averaged F measure, for which relatively fewer solutions exist, and shall develop novel maximization algorithms under both approaches.
2016
Inglese
61
394
404
11
Esperti anonimi
internazionale
scientifica
Multi-label classification; F measure; Learning algorithms; Empirical utility maximization; Decision-theoretic approach
no
Pillai, Ignazio; Fumera, Giorgio; Roli, Fabio
1.1 Articolo in rivista
info:eu-repo/semantics/article
1 Contributo su Rivista::1.1 Articolo in rivista
262
3
partially_open
Files in This Item:
File Size Format  
PattRec 2017a.pdf

Solo gestori archivio

Description: Articolo principale
Type: versione editoriale
Size 647.94 kB
Format Adobe PDF
647.94 kB Adobe PDF & nbsp; View / Open   Request a copy
paper.pdf

open access

Description: Articolo principale
Type: versione post-print
Size 804.99 kB
Format Adobe PDF
804.99 kB Adobe PDF View/Open

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

Questionnaire and social

Share on:
Impostazioni cookie