Evaluation Protocol of Early Classifiers over Multiple Data Sets

Dachraoui, Asma; Bondu, Alexis; Cornuéjols, Antoine

doi:10.1007/978-3-319-12640-1_66

Cited by 4 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If we take that value as a worst-case scenario we can conduct several statistical tests to show the significance of the obtained results. In particular, we applied the Friedman and Nemenyi statistics in order to look for statistical significance among the obtained performances [20]. In order to compare the performances obtained for each of the three methods on the two datasets, Table 4 shows the mean rank of each method considering the 14 different scores, corresponding to the different label performances.…”

Section: Resultsmentioning

confidence: 99%

Spatial codification of label predictions in multi‐scale stacked sequential learning: a case study on multi‐class medical volume segmentation

Sampedro

Escalera

2015

IET Computer Vision

View full text Add to dashboard Cite

In this study, the authors propose the spatial codification of label predictions within the multi-scale stacked sequential learning (MSSL) framework, a successful learning scheme to deal with non-independent identically distributed data entries. After providing a motivation for this objective, they describe its theoretical framework based on the introduction of the blurred shape model as a smart descriptor to codify the spatial distribution of the predicted labels and define the new extended feature set for the second stacked classifier. They then particularise this scheme to be applied in volume segmentation applications. Finally, they test the implementation of the proposed framework in two medical volume segmentation datasets, obtaining significant performance improvements (with a 95% of confidence) in comparison to standard Adaboost classifier and classical MSSL approaches.

show abstract

Section: Resultsmentioning

confidence: 99%

Spatial codification of label predictions in multi‐scale stacked sequential learning: a case study on multi‐class medical volume segmentation

Sampedro

Escalera

2015

IET Computer Vision

View full text Add to dashboard Cite

show abstract

“…• Boosting (AB): este método separa conjuntos de dados reamostrados onde diferente de Bagging, o nível de importância do voto neste método é baseado no desempenho de cada modelo, ao invés de todos os conjuntos terem o mesmo peso (Batista, 2022). Nemenyi, que realiza uma observação pareada de um conjunto de séries de dados e aponta qual a diferença estatística entre elas (Demšar, 2006).…”

Section: Processo De Aprendizadounclassified

Utilização de aprendizado de máquina para classificação de perfis de consumidores de energia elétrica nas regiões do Brasil

Siqueira,

Canuto,

Junior

2023

Cuad. Ed. Desar.

View full text Add to dashboard Cite

A previsão eficiente do consumo de energia pode contribuir significativamente para melhorar a gestão da distribuição e, potencialmente, contribuir para melhorar o planejamento da demanda, melhorar a eficiência energética, controle tarifário e prevenção de falhas. Por outro lado, os avanços nas técnicas computacionais baseadas em dados estão se tornando cada vez mais robustos e populares à medida que alcançam boa precisão nos resultados. Dentre essas, as técnicas de Aprendizado de Máquina (AM) têm sido amplamente utilizadas em diversos domínios de aplicação. Este estudo propõe o desenvolvimento de um modelo capaz de classificar perfis de consumidores de energia, utilizando técnicas de Aprendizado de Máquina. A aplicação dessas técnicas na predição de consumo de energia pode indicar grande potencial para controlar e gerenciar a produção e distribuição de energia elétrica, o que pode trazer maior eficiência, melhorar a produção e otimizar a distribuição. Neste estudo, combinamos uma metodologia de AM, a transferência de aprendizado que é capaz de utilizar conhecimentos pré-estabelecidos em novos contextos, tornando o processo de previsão de energia mais eficiente e robusto. A aplicação da transferência de aprendizado apresentou como resultado acurácias médias acima de 90% nos métodos Bagging, Boosting, Random Forest para todas as bases utilizadas como alvos de transferência.

show abstract

“…One of most useful technique for statistical evaluation of an early classifier is proposed in [78]. As early classifiers address two conflicting objectives (i.e., earliness and accuracy) together, comparing the statistical significance of one early classifier with other becomes more challenging.…”

Section: Statistical Evaluation Of Early Classifiermentioning

confidence: 99%

“…As early classifiers address two conflicting objectives (i.e., earliness and accuracy) together, comparing the statistical significance of one early classifier with other becomes more challenging. In [78], the authors therefore employed two well known statistical methods including Wilcoxon signed-rank test [79] and Pareto optimum [80], for evaluating the early classifiers on many UCR datasets [19]. The evaluation technique uses Wilcoxon signedrank test for independent comparison where it compares two early classifiers on both objectives independently on same dataset.…”

Section: Statistical Evaluation Of Early Classifiermentioning

confidence: 99%

Approaches and Applications of Early Classification of Time Series: A Review

Gupta,

Biswas

et al. 2020

Preprint

View full text Add to dashboard Cite

Early classification of time series has been extensively studied for minimizing class prediction delay in timesensitive applications such as healthcare and finance. A primary task of an early classification approach is to classify an incomplete time series as soon as possible with some desired level of accuracy. Recent years have witnessed several approaches for early classification of time series. As most of the approaches have solved the early classification problem with different aspects, it becomes very important to make a thorough review of the existing solutions to know the current status of the area. These solutions have demonstrated reasonable performance in a wide range of applications including human activity recognition, gene expression based health diagnostic, industrial monitoring, and so on. In this paper, we present a systematic review of current literature on early classification approaches for both univariate and multivariate time series. We divide various existing approaches into four exclusive categories based on their proposed solution strategies. The four categories include prefix based, shapelet based, model based, and miscellaneous approaches. The authors also discuss the applications of early classification in many areas including industrial monitoring, intelligent transportation, and medical. Finally, we provide a quick summary of the current literature with future research directions.

show abstract

Evaluation Protocol of Early Classifiers over Multiple Data Sets

Cited by 4 publications

References 7 publications

Spatial codification of label predictions in multi‐scale stacked sequential learning: a case study on multi‐class medical volume segmentation

Spatial codification of label predictions in multi‐scale stacked sequential learning: a case study on multi‐class medical volume segmentation

Utilização de aprendizado de máquina para classificação de perfis de consumidores de energia elétrica nas regiões do Brasil

Approaches and Applications of Early Classification of Time Series: A Review

Contact Info

Product

Resources

About