A convex hull-based data selection method for data driven models

Khosravani, H.; Ruano, A. E.; Ferreira, Pedro

doi:10.1016/j.asoc.2016.06.014

Cited by 30 publications

(19 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This work uses the ApproxHull algorithm proposed in [35], to select data for training, testing and validation data for the artificial neural networks design. ApproxHull is an incremental randomized approximate convex hull (CH) algorithm that selects the points involving whole data points.…”

Section: Data Set Constructionmentioning

confidence: 99%

“…Then, it generates a population of k facets based on the existing convex hull, selecting the furthest points in the current facets' population as new vertices of the convex hull, which are integrated into the current convex hull. A detailed explanation of the convex hull algorithm may be found in [35].…”

Section: Data Set Constructionmentioning

confidence: 99%

“…For this reason, the structure of the chromosome includes the number of neurons and pointers for the admissible delays of the endogenous and exogeneous variable(s), if available. Details on the MOGA method were previously published in [35]. Before being evaluated in MOGA, each model has its parameters determined by a Levenberg-Marquardt (LM) algorithm [36,37] minimizing an error criterion that exploits the linear-nonlinear relationship of the RBF NN model parameters [38,39].…”

Section: Moga Designmentioning

confidence: 99%

See 2 more Smart Citations

Short-Term Forecasting Photovoltaic Solar Power for Home Energy Management Systems

Bot

Ruano

2021

Inventions

Self Cite

View full text Add to dashboard Cite

Accurate photovoltaic (PV) power forecasting is crucial to achieving massive PV integration in several areas, which is needed to successfully reduce or eliminate carbon dioxide from energy sources. This paper deals with short-term multi-step PV power forecasts used in model-based predictive control for home energy management systems. By employing radial basis function (RBFs) artificial neural networks (ANN), designed using a multi-objective genetic algorithm (MOGA) with data selected by an approximate convex-hull algorithm, it is shown that excellent forecasting results can be obtained. Two case studies are used: a special house located in the USA, and the other a typical residential house situated in the south of Portugal. In the latter case, one-step-ahead values for unscaled root mean square error (RMSE), mean relative error (MRE), normalized mean average error (NMAE), mean absolute percentage error (MAPE) and R2 of 0.16, 1.27%, 1.22%, 8% and 0.94 were obtained, respectively. These results compare very favorably with existing alternatives found in the literature.

show abstract

Section: Data Set Constructionmentioning

confidence: 99%

Section: Data Set Constructionmentioning

confidence: 99%

Section: Moga Designmentioning

confidence: 99%

See 1 more Smart Citation

Short-Term Forecasting Photovoltaic Solar Power for Home Energy Management Systems

Bot

Ruano

2021

Inventions

Self Cite

View full text Add to dashboard Cite

show abstract

“…To enable MOGA to generate models applicable to the whole range of data where the classifier is going to be used, we included all convex points [36] of _ into the training set. To obtain the convex points, the Approxhull algorithm [37,38] is used, resulting in 13023 samples, among which 11732 were normal and 1291 abnormal. The convex points along with 6977 random data samples (50% normal and 50% abnormal) constitute our training set whose size is 20,000.…”

Section: Constructing the Input Dataset For Mogamentioning

confidence: 99%

An intelligent support system for automatic detection of cerebral vascular accidents from brain CT images

Hajimani

Ruano

2017

Computer Methods and Programs in Biomedicine

Self Cite

View full text Add to dashboard Cite

Values of specificity of 98% and sensitivity of 98% were obtained, at pixel level, by an ensemble of non-dominated models generated by MOGA, in a set of 150 CT slices (1,867,602pixels), marked by a NeuroRadiologist. This approach also compares favorably at a lesion level with three other published solutions, in terms of specificity (86% compared with 84%), degree of coincidence of marked lesions (89% compared with 77%) and classification accuracy rate (96% compared with 88%).

show abstract

“…Em seleção de observações, o objetivo é selecionar os dados mais importantes do conjunto de treinamento para uma determinada tarefa ou análise. Essa área vem sendo amplamente estudada, uma vez que suas aplicações podem ser vistas em sistemas de classificação de imagens, utilizando Active Learning [8], [9], problemas de otimização dinâmica, com Query Based Learning [10] e ainda pesquisas para se desenvolver melhores classificadores, como por exemplo o DROP3 [11].…”

unclassified

Seleção de Observações Aplicada à Classificação

Baruque¹,

Estrada²,

Peres³

2020

Anais Do 14. Congresso Brasileiro De Inteligência Computacional

View full text Add to dashboard Cite

Atualmente, técnicas na área de aprendizado estatístico vêm sendo amplamente utilizadas tanto em pesquisas científicas quanto em aplicações no mercado e indústria. Em função da grande quantidade de dados gerada com a informatização de sistemas, aplicações em Big Data demandam desenvolver e aprimorar metodologias capazes de extrair informações úteis a partir de grandes conjuntos de dados. Este artigo tem como proposta o desenvolvimento de um método de seleção de observações do conjunto de treinamento. O objetivo é verificar se é possível manter o desempenho de classificação utilizando quantidades cada vez menores de dados. O algoritmo utiliza uma estimativa do classificador Bayesiano para determinar as observações que serão selecionadas. Este procedimento foi aplicado ao algoritmo de vizinhos mais próximos (KNN) e aplicações foram realizadas em três bancos de dados, com bom desempenho do método. Os resultados se mostraram positivos em manter e até melhorar a classificação utilizando as observações selecionadas.

show abstract

A convex hull-based data selection method for data driven models

Cited by 30 publications

References 42 publications

Short-Term Forecasting Photovoltaic Solar Power for Home Energy Management Systems

Short-Term Forecasting Photovoltaic Solar Power for Home Energy Management Systems

An intelligent support system for automatic detection of cerebral vascular accidents from brain CT images

Seleção de Observações Aplicada à Classificação

Contact Info

Product

Resources

About