arrow
Return

Data complexity meta-features for regression problems

delete2017-12-21
delete42
delete
OA
AI
A
Ana Carolina Lorena *
A
Aron I. Maciel
P
Péricles Miranda
I
Ivan G. Costa
R
Ricardo B. C. Prudêncio
DOI:10.1007/s10994-017-5681-1delete
deleteOriginal
deleteShare
deleteSave
View PDF
Abstract

Abstract

En 中文
In meta-learning, classification problems can be described by a variety of features, including complexity measures. These measures allow capturing the complexity of the frontier that separates the classes. For regression problems, on the other hand, there is a lack of such type of measures. This paper presents and analyses measures devoted to estimate the complexity of the function that should fitted to the data in regression problems. As case studies, they are employed as meta-features in three meta-learning setups: (i) the first one predicts the regression function type of some synthetic datasets; (ii) the second one is designed to tune the parameter values of support vector regressors; and (iii) the third one aims to predict the performance of various regressors for a given dataset. The results show the suitability of the new measures to describe the regression datasets and their utility in the meta-learning tasks considered. In cases (ii) and (iii) the achieved results are also similar or better than those obtained by the use of classical meta-features in meta-learning.
Keywords:
Meta-learning
Meta-features
Complexity measures

Journal

Machine Learning cover
Machine Learning
IF:
2.9
Papers:
2.8K
Citations:
3.8W

Organization

R
RWTH Aachen University
Scholars:
2.4K
Papers: 601
Citations: 2.0W
U
universidade federal de são paulo (unifesp)
Scholars:
0
Papers: 1
Citations: 6.1K
U
universidade federal de pernambuco
Scholars:
434
Papers: 136
Citations: 2.4K
researcher View more organizations
Cited Papers

Cited Papers

An update on therapeutics for tobacco dependence
err2007-12-13
err0
PREAI
errTim Lancaster; Lindsay Stead; Kate Cahill
errShare
errSave
Stillbirth classification in population-based data and role of fetal growth restriction: the example of RECODE
err2013-10-03
err0
errOAAI
errAnne Ego; Jennifer Zeitlin; Pierre Batailler; Séverine Cornec; Anne Fondeur; Marion Baran-Marszak; Pierre-Simon Jouk; Thierry Debillon; Christine Cans
errShare
errSave
The influence of time of rheological parameters of fresh cement slurries
err2014-01-01
err0
errOAAI
errStanisław Stryczek; Rafał Wiśniowski; Andrzej Gonet; Albert Złotkowski
errShare
errSave
Noise detection in the meta-learning level
err2016-02-01
err48
PREAI
errGarcia, Luis P. F.; de Carvalho, Andre C. P. L. F.; Lorena, Ana C.
errShare
errSave
Effect of label noise in the complexity of classification problems
err2015-07-01
err101
PREAI
errGarcia, Luis P. F.; de Carvalho, Andre C. P. L. F.; Lorena, Ana C.
errShare
errSave
researcher View more