Reducing noise impact on MLP training

Kordos, Mirosław; Rusiecki, Andrzej

doi:10.1007/s00500-015-1690-9

Cited by 27 publications

(13 citation statements)

References 25 publications

(49 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is to be seen that BPNN-NAR is a feedforward neural system write demonstrate, while BPNN-NARMA is an intermittent neural system compose display [18]. This investigation reports the blunder measures on the test dataset, which is the most key trademark, mirroring ANN's speculation capacity [19,20]. Table 2 demonstrates the information parceling for organize pre-handling.…”

Section: Methodsmentioning

confidence: 99%

Fitting Conventional Neural Network Time Series Models on Sand Price Indices Dataset

Ghani¹,

Kamaruddin²,

Musirin

et al. 2018

IJET

View full text Add to dashboard Cite

This result-based paper discusses on the best aftereffects of both fitted BPNN-NAR and BPNN-NARMA on MCCI Sand dataset regarding distinctive error measures. This exploration examine the outcomes as far as the execution of the fitted forecasting models by every arrangement of input lags and error lags utilized, the execution of the fitted anticipating models by various hidden nodes ut ilized, the execution of the fitted estimating models when joining both inputs and hidden nodes, the consistency of error measures utilized for the fitted determining models, and in addition the general best fitted estimating models for Malaysian sand price indices dataset. In this examination, Malaysian sand price indices monthly data from January 1980 to December 2013 were adapted. The examination of BPNN-NAR on Malaysian sand data shows that insufficient or inadequate combination of input and error lags lead to greater RMSE. Correspondingly, the number of input lags for BPNN-NAR, as well as the number of input and error lags for BPNN-NARMA really have direct effect to the models' performances. The higher or the lesser hidden nodes to the input lags, the higher the network's RMSE. On the other hand, the higher the input lags lead to the higher network's RMSE.

show abstract

Section: Methodsmentioning

confidence: 99%

Fitting Conventional Neural Network Time Series Models on Sand Price Indices Dataset

Ghani¹,

Kamaruddin²,

Musirin

et al. 2018

IJET

View full text Add to dashboard Cite

show abstract

“…In this context, Jorgensen and Duffy stated that the accuracy of a model cannot exceed the accuracy of the experimental data [6]. Although this statement is correct, it can further be consolidated since machine learning (ML) algorithms are capable of dealing with errors in the training data [11]. To put it differently, the observed performance of a model cannot be better than the internal error of the test set.…”

Section: • Training and Testing The Modelmentioning

confidence: 99%

Pushing the limits of solubility prediction via quality-oriented data selection

Sorkun

Koelman

2020

Preprint

View full text Add to dashboard Cite

Accurate prediction of the solubility of chemical substances in solvents remains a challenge. The sparsity of high-quality solubility data is recognized as the biggest hurdle in the development of robust data-driven methods for practical use. Nonetheless, the effects of the quality and quantity of data on aqueous solubility predictions have not yet been scrutinized. In this study, the roles of the size and the quality of datasets on the performances of the solubility prediction models are unraveled, and the concepts of actual and observed performances are introduced. In an effort to curtail the gap between actual and observed performances, a quality-oriented data selection method, which evaluates the quality of data and extracts the most accurate part of it through statistical validation, is designed. Applying this method on the largest publicly available solubility database and using a consensus machine learning approach, a top-performing solubility prediction model is achieved.

show abstract

“…It is also worth mentioning that the use of instance selection methods is not limited to the nearest neighbor algorithm, but has also found applications to many other predictors. For example, Kordos and Rusiecki (2016) evaluated various approaches to MLP neural network training in the presence of noise. Their results pointed out that the ENN was among the best methods.…”

Section: 1mentioning

confidence: 99%

Ensembles of instance selection methods: A comparative study

Blachnik

2019

International Journal of Applied Mathematics and Computer Science

View full text Add to dashboard Cite

Instance selection is often performed as one of the preprocessing methods which, along with feature selection, allows a significant reduction in computational complexity and an increase in prediction accuracy. So far, only few authors have considered ensembles of instance selection methods, while the ensembles of final predictive models attract many researchers. To bridge that gap, in this paper we compare four ensembles adapted to instance selection: Bagging, Feature Bagging, AdaBoost and Additive Noise. The last one is introduced for the first time in this paper. The study is based on empirical comparison performed on 43 datasets and 9 base instance selection methods. The experiments are divided into three scenarios. In the first one, evaluated on a single dataset, we demonstrate the influence of the ensembles on the compression–accuracy relation, in the second scenario the goal is to achieve the highest prediction accuracy, and in the third one both accuracy and the level of dataset compression constitute a multi-objective criterion. The obtained results indicate that ensembles of instance selection improve the base instance selection algorithms except for unstable methods such as CNN and IB3, which is achieved at the expense of compression. In the comparison, Bagging and AdaBoost lead in most of the scenarios. In the experiments we evaluate three classifiers: 1NN, kNN and SVM. We also note a deterioration in prediction accuracy for robust classifiers (kNN and SVM) trained on data filtered by any instance selection methods (including the ensembles) when compared with the results obtained when the entire training set was used to train these classifiers.

show abstract

Reducing noise impact on MLP training

Cited by 27 publications

References 25 publications

Fitting Conventional Neural Network Time Series Models on Sand Price Indices Dataset

Fitting Conventional Neural Network Time Series Models on Sand Price Indices Dataset

Pushing the limits of solubility prediction via quality-oriented data selection

Ensembles of instance selection methods: A comparative study

Contact Info

Product

Resources

About