A Preliminary Study of Ordinal Metrics to Guide a Multi-Objective Evolutionary Algorithm

Research areas:

• Methodology - Machine learning

Year:

2011

Type of Publication:

In Proceedings

Keywords:

Mean Absolute Error, Multi-Objective Evolutionary Algorithm, Ordinal Measures

Authors:

• Cruz-Ramírez, Manuel
• Hervás-Martínez, César
• Sánchez-Monedero, Javier
• Gutiérrez, Pedro Antonio

Book title:

11th International Conference on Intelligent Systems Design andApplications (ISDA 2011)

Pages:

1176-1181

Address:

Cordoba, Spain, Spain

Month:

November

BibTex:

@conference{Cruz2011isda, author = "Manuel Cruz-Ram{\'i}rez and C{\'e}sar Herv{\'a}s-Mart{\'i}nez and Javier S{\'a}nchez-Monedero and Pedro Antonio Guti{\'e}rrez", abstract = "There are many metrics available to measure the goodness of a classifier when working with ordinal datasets. These measures are divided into product-moment and association metrics. In this paper, the behavior of several metrics is studied in different situations. In addition, two new measures associated with an ordinal classifier are defined: the maximum and the minimum mean absolute error of all the classes. From the results of this comparison, a pair of metrics is selected (one associated to the overall error and another one to the error of the class with lowest level of classification) to guide the evolution of a multi-objective evolutionary algorithm, obtaining good results in generalization on ordinal datasets.", address = "Cordoba, Spain, Spain", booktitle = "11th International Conference on Intelligent Systems Design andApplications (ISDA 2011)", keywords = "Mean Absolute Error, Multi-Objective Evolutionary Algorithm, Ordinal Measures", month = "nov", pages = "1176-1181", title = "{A} {P}reliminary {S}tudy of {O}rdinal {M}etrics to {G}uide a {M}ulti-{O}bjective {E}volutionary {A}lgorithm", year = "2011", }

Abstract:

There are many metrics available to measure the goodness of a classifier when working with ordinal datasets. These measures are divided into product-moment and association metrics. In this paper, the behavior of several metrics is studied in different situations. In addition, two new measures associated with an ordinal classifier are defined: the maximum and the minimum mean absolute error of all the classes. From the results of this comparison, a pair of metrics is selected (one associated to the overall error and another one to the error of the class with lowest level of classification) to guide the evolution of a multi-objective evolutionary algorithm, obtaining good results in generalization on ordinal datasets.