Bagging classification tree-based robust variable selection for radial basis function network modeling in metabonomics data analysis

Gu, Hui; Cui, Yanfang; Xu, Lu; Tu, Meng-Ying; Fu, Yan-Jiao; Fu, Haiyan; Zhou, Yanping

doi:10.1016/j.chemolab.2017.12.014

Cited by 9 publications

(5 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Boosting [ 14 ] and bagging [ 15 ] are two popular combined classification methods. Among them, bagging is a typical representative of a parallel ensemble learning method [ 16 ].…”

Section: Introductionmentioning

confidence: 99%

Cancer Classification with a Cost-Sensitive Naive Bayes Stacking Ensemble

Xiong

2021

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

Ensemble learning combines multiple learners to perform combinatorial learning, which has advantages of good flexibility and higher generalization performance. To achieve higher quality cancer classification, in this study, the fast correlation-based feature selection (FCBF) method was used to preprocess the data to eliminate irrelevant and redundant features. Then, the classification was carried out in the stacking ensemble learner. A library for support vector machine (LIBSVM), K -nearest neighbor (KNN), decision tree C4.5 (C4.5), and random forest (RF) were used as the primary learners of the stacking ensemble. Given the imbalanced characteristics of cancer gene expression data, the embedding cost-sensitive naive Bayes was used as the metalearner of the stacking ensemble, which was represented as CSNB stacking. The proposed CSNB stacking method was applied to nine cancer datasets to further verify the classification performance of the model. Compared with other classification methods, such as single classifier algorithms and ensemble algorithms, the experimental results showed the effectiveness and robustness of the proposed method in processing different types of cancer data. This method may therefore help guide cancer diagnosis and research.

show abstract

“…Boosting [ 14 ] and bagging [ 15 ] are two popular combined classification methods. Among them, bagging is a typical representative of a parallel ensemble learning method [ 16 ].…”

Section: Introductionmentioning

confidence: 99%

Cancer Classification with a Cost-Sensitive Naive Bayes Stacking Ensemble

Xiong

2021

Computational and Mathematical Methods in Medicine

View full text Add to dashboard Cite

show abstract

“…For the experimental results, amazon product review data from Kaggle [22] is tested and as well as tested against the realtime data obtained from the amazon product URL. Experimental results are compared with traditional models such as Adaboost with Random tree [16], KNN [18], Stacking Random tree [19], Bagging Random tree [20], Naïve Bayes Multinomial Text [21], RF [17] for performance analysis. The main problems identified in these models are 1.…”

Section: Resultsmentioning

confidence: 99%

A Filter Based Improved Decision Tree Sentiment Classification Model for RealTime Amazon Product Review Data

Syamala¹,

Nalini²

2020

IJIES

View full text Add to dashboard Cite

E-Commerce product features and reviews are considered to be the essential factors in real-time e-commerce sites for product recommendation systems. Due to inaccuracy decision patterns, in most cases e-commerce user fails to predict the products based on the user ratings and review comments. Traditional sentiment classification models are independent of data filtering, transformation and sentiment score computing techniques which require high computing memory, time and mostly leading to false-positive rate. To overcome these issues, a novel sentiment score-based product recommendation model is proposed on the real-time product data. In this model, a new product ranking score, filtering, and hybrid decision tree classifiers are implemented. Initially, real-time amazon product review data is captured using Document Object Model (DOM) parser. The features from the review comments are extracted using lexicon Feature Dictionary (FD) and AFINN, Normalized Product Review Score (NPRS) are generated to compute the class label for product review sentiment prediction. Ranked Principal Component Analysis (RPCA) is used as a feature selection measure to overcome the problem of data sparsity. Random Tree, Hoeffding Tree, Adaboost + Random Tree, the three variants of decision tree classifiers are used for product sentiment classification. The proposed filter-based improved decision tree sentiment classification model for real-time amazon product review data recommends the product based on the user query by prediction using a new novel normalized product review sentiment score and ranked feature selection measure. The proposed product recommendation, the decision-making system maximizes sentiment classification accuracy. Experimental results are compared against the traditional decisionmaking classification models in terms of correctly classified instances, error rate, and PRC, F-measure, kappa statistics. The proposed model experimental results show high efficiency.

show abstract

“…Combination of more models is going to be a task for further research. Nowadays combination of models on various theoretical bases is a theme of a lot of papers [1][2][3][4][5]. In further research variance of optimal combination of forecasters is going to be investigated in case of ARIMA (p, d, q) and GARCH (p, q) time series models.…”

Section: Discussionmentioning

confidence: 99%

“…In expression (3) i  are standard deviations of the models,  is correlation between forecast errors of these models [10,11,14]. Variance of various combinations is thoroughly described in [14].…”

Section: Variance Of Ardl Models Forecasts and Forecast Combinationmentioning

confidence: 99%

“…The simplest combination function in case of time series processing is just averaging of all models' predictions or selection of the best one [1]. Approaches to implementation of the bagging and boosting principles [2,3] in order to construct better time series models than original ones (and usually implemented in time series analysis) have already been published [4][5][6][7]. But question whether boosting or bagging can be used in models of connections between pairs of time series hasn't been observed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reduction of time series forecast variance by the case of ArDL(p, q) models

Petrusevich

2022

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Combination of time series forecasts or time series models is often considered nowadays in various research papers and is usually a good technique in practice. Still it has got weak theoretical explanation and there’s a lot of work that is going to be done in this field. In the research optimal combination of forecasters in terms of minimal variance is investigated. These results don’t depend on nature of combined models but in the practical part ArDL models describing connections between pairs of time series are considered. Weighted sum of models’ predictions is used as a new prediction with variance that should be less than individual models’ predictions variance. Optimal weights for combination of two models are calculated theoretically. They are based on correlation of combined models and their forecasts’ variance. Optimal model combination variance (variance of the combined models with optimal weights) has been investigated and it has been shown that it doesn’t exceed variances of the combined models. Practical experiments and conditions under which combined time series model is expected to improve forecasts of original ones are in scope of this research. Forecasts of two ArDL models are combined in order to construct forecast with lower variety. This approach can be used in case of more models but theoretical expression of optimal weights is going to be a part of further research. Russian macroeconomical time series statistics and economic data of the USA are used as experimental time series. Combined models have got predictions close to the forecasts of the best models and better than the worst models used in combinations. It’s shown that optimal combination should make predictions of high quality if combined predictors describe different aspects of investigated time series and correlation of their forecasts isn’t very high. Otherwise, it turns into mean forecast of two models.

show abstract

Bagging classification tree-based robust variable selection for radial basis function network modeling in metabonomics data analysis

Cited by 9 publications

References 33 publications

Cancer Classification with a Cost-Sensitive Naive Bayes Stacking Ensemble

Cancer Classification with a Cost-Sensitive Naive Bayes Stacking Ensemble

A Filter Based Improved Decision Tree Sentiment Classification Model for RealTime Amazon Product Review Data

Reduction of time series forecast variance by the case of ArDL(p, q) models

Contact Info

Product

Resources

About