Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ LAReferencia - Red F...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
versions View all 1 versions
addClaim

Comparação de técnicas de aprendizado de máquina para a classificação da qualidade da madeira

Authors: Roder, Mateus; Rossi, André Luis Debiaso; Affonso, Carlos de Oliveira;

Comparação de técnicas de aprendizado de máquina para a classificação da qualidade da madeira

Abstract

A classificação da qualidade da madeira é uma tarefa crucial para que as empresas fornecedoras dessa matéria prima logrem êxito no seu desenvolvimento econômico. Essa classificação é realizada, geralmente, por especialistas, o que torna essa tarefa subjetiva e de alto custo. Assim, buscou-se na computação, em específico na área de Aprendizado de Máquina (AM), soluções automatizadas com altas taxas de acerto e maior eficiência. Atualmente, diversas técnicas de AM estão disponíveis e podem ser usadas para esse propósito. Porém, escolher o melhor algoritmo para o problema sob análise não é uma tarefa trivial. Sendo assim, o objetivo deste artigo é comparar diferentes algoritmos de AM para o problema de classificação da qualidade da madeira. Para isso, realizaram-se experimentos usando três algoritmos de AM: Redes Neurais Artificiais, Máquinas de Vetores de Suporte e Árvores de Decisão. Esses algoritmos representam diferentes vieses de classificação. Um total de 374 instâncias obtidas a partir da análise de imagens de tábuas de madeira foram utilizadas nessa avaliação. Cada instância é descrita pelas características da imagem e por um atributo alvo, que identifica a qualidade da madeira. A partir das análises dos resultados, observou-se que os desempenhos preditivos dos algoritmos foram similares, com pequena vantagem para a máquina de vetores de suporte.

The wood quality classification is a crucial task for the suppliers of this raw material to succeed in their economic development. This classification is usually performed by experts, which makes this task subjective and expensive. Thus, computing tools have been developed, in particular in machine learning (ML) area, to accomplish this task efficiently and with high accuracy rates. Currently, several ML techniques are available and can be employed for this purpose. However, choosing which is the best algorithm for the problem under analysis is not a trivial task. Therefore, the purpose of this article is to compare different ML techniques for the wood quality classification problem. For such, we carried out experiments using three ML algorithms: Artificial Neural Networks, Support Vector Machines and Decision Trees. These algorithms represent different search bias. An amount of 374 instances obtained from wood boards images were used for training and test of the models. Each instance is described by the features of an image and a target attribute, which identifies the quality of this timber. The results showed that the predictive performance of the algorithms were similar, with slightly superiority of the support vector machines.

Pró-Reitoria de Extensão Universitária (PROEX UNESP)

Country
Brazil
Keywords

Wood quality, Mineração de dados, Classification, Classificação, Data mining, Qualidade da madeira

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green