Building selective ensembles of Randomization Based Neural Networks with the successive projections algorithm

Mesquita, Diego; Gomes, João P. P.; Rodrigues, Leonardo Ramos; Oliveira, Saulo A. F.; Galvão, Roberto Kawakami Harrop

doi:10.1016/j.asoc.2017.08.007

Cited by 30 publications

(11 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SPA is a forward-loop-selecting method which can minimise vector space collinearity. [20] Assuming that the spectral matrix is X nÂp , where n is the number of samples, p is the number of wavelengths. In order to select the optimal bands, the value of kth column in spectral matrix of training set was assigned to x kð1Þ ; k 2 ð1; 2; Λ; pÞbefore the first iteration (n = 1).…”

Section: Characteristic Wavelength Filtering Algorithmmentioning

confidence: 99%

Prediction and visualisation of S-ovalbumin content in egg whites using hyperspectral images

Wang

et al. 2019

International Journal of Food Properties

View full text Add to dashboard Cite

This study proposed a method using hyper-spectral imaging technology in determining eggs' quality in term of freshness from a biochemical perspective by estimating the S-ovalbumin content. This method has the potential in assessing eggs' quality rapidly and non-destructively. Hyper-spectral image of egg was captured using a hyper-spectral imaging system and regression model was built to estimate the S-ovalbumin content. The successive projections algorithm (SPA) was used to select significant wavebands followed by building a partial least squares regression (PLSR) model and a multiple linear regression (MLR) model. The MLR model could predict S-ovalbumin content better than PLSR model with a higher correlation coefficient (0.922) and lower root mean square error (0.086) of the calibration set, a higher correlation coefficient (0.911) and lower root mean square error (0.119) of the validation set, and a higher residual predictive deviation (2.348). The regression equation from the MLR model was used to compute each pixel of the image in the validation set and visualisation of S-ovalbumin content distribution in the egg was obtained using pseudocolor image. The findings implied that the proposed hyper-spectral imaging system with the regression model developed has the potential in determining and visualising the eggs' quality.

show abstract

Section: Characteristic Wavelength Filtering Algorithmmentioning

confidence: 99%

Prediction and visualisation of S-ovalbumin content in egg whites using hyperspectral images

Wang

et al. 2019

International Journal of Food Properties

View full text Add to dashboard Cite

show abstract

“…Many research studies have demonstrated the efficiency of ensemble learning, which is an easy technique. In addition, they indicated that ensemble learning realizes higher efficiency than the individual algorithms with the same complexity [27]. Ensemble learning is the combination of the outputs of sets of learning models according to specified rules to obtain a better model than single learners.…”

Section: Ensemble Modelmentioning

confidence: 99%

“…In aspect of designing an efficient new ensemble model method [24][25][26], there are 4 key points to be considered: dataset, based model, combination strategy and method of assigning probability weight. According to previous studies [16,27,28] the researcher tends to not pay much attention to multi-class data sets due to the multi-class classification. There is a complex decision-making class classification results that are difficult to manage [29].…”

Section: Introductionmentioning

confidence: 99%

Cost-sensitive probability for weighted voting in an ensemble model for multi-class classification problems

Rojarath

Songpan

2021

Appl Intell

View full text Add to dashboard Cite

Ensemble learning is an algorithm that utilizes various types of classification models. This algorithm can enhance the prediction efficiency of component models. However, the efficiency of combining models typically depends on the diversity and accuracy of the predicted results of ensemble models. However, the problem of multi-class data is still encountered. In the proposed approach, cost-sensitive learning was implemented to evaluate the prediction accuracy for each class, which was used to construct a cost-sensitivity matrix of the true positive (TP) rate. This TP rate can be used as a weight value and combined with a probability value to drive ensemble learning for a specified class. We proposed an ensemble model, which was a type of heterogenous model, namely, a combination of various individual classification models (support vector machine, Bayes, K-nearest neighbour, naïve Bayes, decision tree, and multi-layer perceptron) in experiments on 3-, 4-, 5- and 6-classifier models. The efficiencies of the propose models were compared to those of the individual classifier model and homogenous models (Adaboost, bagging, stacking, voting, random forest, and random subspaces) with various multi-class data sets. The experimental results demonstrate that the cost-sensitive probability for the weighted voting ensemble model that was derived from 3 models provided the most accurate results for the dataset in multi-class prediction. The objective of this study was to increase the efficiency of predicting classification results in multi-class classification tasks and to improve the classification results.

show abstract

“…In addition, a multiscatter correction (MSC) was performed to eliminate the effect of scattering and enhance the spectral absorption information related to oleic acid and linoleic acid (Chu et al, 2018). In the calculation process, the average spectra of all samples were used as the reference spectrum (Mesquita, Gomes, Rodrigues, Oliveira, & Galvão, 2018). The spectra of all other samples were corrected according to the reference spectrum, including baseline translation and offset correction.…”

Section: Spectral Data Preprocessingmentioning

confidence: 99%

Nondestructive phenotyping fatty acid trait of single soybean seeds using reflective hyperspectral imagery

Zhou

Scaboo

et al. 2021

J Food Process Engineering

View full text Add to dashboard Cite

Soybean (Glycine max) is one of the most economically important crops in the world and is used widely for different purposes. Breeding programs have developed new varieties with desired traits, including altered fatty acid profiles and levitated protein content. The breeding process involves the selection of elite genotypes from a large number of experimental lines, which is time-consuming and costly. This study aimed to evaluate the feasibility of a reflective hyperspectral imaging system in classifying the fatty acid content of single soybean seeds for high-throughput phenotyping high oleic soybean cultivars. A hyperspectral imager was used to take hyperspectral images of 100 soybean seeds of two varieties. After comparing the performance of

show abstract

Building selective ensembles of Randomization Based Neural Networks with the successive projections algorithm

Cited by 30 publications

References 27 publications

Prediction and visualisation of S-ovalbumin content in egg whites using hyperspectral images

Prediction and visualisation of S-ovalbumin content in egg whites using hyperspectral images

Cost-sensitive probability for weighted voting in an ensemble model for multi-class classification problems

Nondestructive phenotyping fatty acid trait of single soybean seeds using reflective hyperspectral imagery

Contact Info

Product

Resources

About