scholarly journals Subset selection of markers for the genome-enabled prediction of genetic values using radial basis function neural networks

2020 ◽  
Vol 43 ◽  
pp. e46307 ◽  
Author(s):  
Isabela de Castro Sant'Anna ◽  
Gabi Nunes Silva ◽  
Moysés Nascimento ◽  
Cosme Damião Cruz
Keyword(s):  
Neural Networks ◽  
Root Mean Square Error ◽  
Radial Basis Function ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Basis Function ◽  
Stepwise Regression ◽  
Subset Selection ◽  
Mean Square ◽  
Selection Of

This paper aimed to evaluate the effectiveness of subset selection of markers for genome-enabled prediction of genetic values using radial basis function neural networks (RBFNN). To this end, an F1 population derived from the hybridization of divergent parents with 500 individuals genotyped with 1000 SNP-type markers was simulated. Phenotypic traits were determined by adopting three different gene action models – additive, additive-dominant, and epistatic, representing two dominance situations: partial and complete with quantitative traits having a heritability (h2) of 30 and 60%; traits were controlled by 50 loci, considering two alleles per locus. Twelve different scenarios were represented in the simulation. The stepwise regression was used before the prediction methods. The reliability and the root mean square error were used for estimation using a fivefold cross-validation scheme. Overall, dimensionality reduction improved the reliability values for all scenarios, specifically with h2 =30 the reliability value from 0.03 to 0.59 using RBFNN and from 0.10 to 0.57 with RR-BLUP in the scenario with additive effects. In the additive dominant scenario, the reliability values changed from 0.12 to 0.59 using RBFNN and from 0.12 to 0.58 with RR-BLUP, and in the epistasis scenarios, the reliability values changed from 0.07 to 0.50 using RBFNN and from 0.06 to 0.47 with RR-BLUP. The results showed that the use of stepwise regression before the use of these techniques led to an improvement in the accuracy of prediction of the genetic value and, mainly, to a large reduction of the root mean square error in addition to facilitating processing and analysis time due to a reduction in dimensionality.

scholarly journals Prediction Algorithm of Parameters of Toe Clearance in the Swing Phase

2019 ◽  
Vol 2019 ◽  
pp. 1-10
Author(s):  
Tamon Miyake ◽  
Masakatsu G. Fujie ◽  
Shigeki Sugano
Keyword(s):  
Root Mean Square Error ◽  
Radial Basis Function ◽  
Mean Square Error ◽  
Basis Function ◽  
Radial Basis Function Network ◽  
Swing Phase ◽  
Mean Square ◽  
Data Points ◽  
Predicted Values ◽  
Function Network

The adaptive control of gait training robots is aimed at improving the gait performance by assisting motion. In conventional robotics, it has not been possible to adjust the robotic parameters by predicting the toe motion, which is considered a tripping risk indicator. The prediction of toe clearance during walking can decrease the risk of tripping. In this paper, we propose a novel method of predicting toe clearance that uses a radial basis function network. The input data were the angles, angular velocities, and angular accelerations of the hip, knee, and ankle joints in the sagittal plane at the beginning of the swing phase. In the experiments, seven subjects walked on a treadmill for 360 s. The radial basis function network was trained with gait data ranging from 20 to 200 data points and tested with 100 data points. The root mean square error between the true and predicted values was 3.28 mm for the maximum toe clearance in the earlier swing phase and 2.30 mm for the minimum toe clearance in the later swing phase. Moreover, using gait data of other five subjects, the root mean square error between the true and predicted values was 4.04 mm for the maximum toe clearance and 2.88 mm for the minimum toe clearance when the walking velocity changed. This provided higher prediction accuracy compared with existing methods. The proposed algorithm used the information of joint movements at the start of the swing phase and could predict both the future maximum and minimum toe clearances within the same swing phase.

scholarly journals Subset selection of markers for genome-enabled prediction of genetic val-ues using radial basis function neural networks

2018 ◽  
Author(s):  
Isabela de Castro Sant' Anna ◽  
Gabi Nunes Silva ◽  
Moysés Nascimento ◽  
Cosme Damiao Cruz
Keyword(s):  
Neural Networks ◽  
Radial Basis Function ◽  
Basis Function ◽  
Mean Squared Error ◽  
Subset Selection ◽  
Predictive Ability ◽  
Phenotypic Traits ◽  
Radial Basis ◽  
The Mean ◽  
Selection Of

This paper aimed to evaluate the efficiency of subset selection of markers for genome-enabled prediction of genetic values using radial basis function neural networks (RBFNN). For this purpose, an F1 population from hybridization of divergent parents with 500 individuals geno-typed with 1,000 SNP-type markers was simulated. Phenotypic traits were determined by adopting three different gene action models – additive, additive-dominant, and epistasic , com-plying with two dominance situations: partial and complete with quantitative traits admitting heritability (h2) equal to 30 and 60%, each one controlled by 50 loci, considering two alleles per locus, totaling 12 different scenarios. To evaluate the predictive ability of RR_BLUP and the neural networks, a cross-validation procedure with five replicates were trained using 80% of the individuals of the population. Two methods were used: dimensionality reduction and stepwise regression. The square of the correlation between the predicted genomic estimated breeding val-ue (GEBV) and the phenotype value was used to measure predictive reliability. For h2 = 0.3 in the additive scenario, the R2 values were 59% for neural network (RBFNN) and 57% for RR-BLUP, and in the epistatic scenario, R2 values were 50% and 41%, respectively. Additionally, when analyzing the mean-squared error root, the difference in performance between the tech-niques is even greater. For the additive scenario, the estimates were 91 for RR-BLUP and 5 for neural networks and, in the most critical scenario, they were 427 for RR-BLUP and 20 for neu-ral network. The results showed that the use of neural networks and variable selection tech-niques allows capturing epistasis interactions, leading to an improvement in the accuracy of pre-diction of the genetic value and, mainly, to a large reduction of the mean square error, which indicates greater genomic value.

scholarly journals Penerapan Generalized Regression Neural Networks untuk Memprediksi Produksi Padi Terhadap Perubahan Iklim

2017 ◽  
Vol 2 (2) ◽  
pp. 117 ◽  
Author(s):  
Muhammad Alkaff ◽  
Yuslena Sari
Keyword(s):  
Neural Networks ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Input Data ◽  
Mean Square ◽  
Generalized Regression Neural Networks ◽  
Data Output ◽  
Generalized Regression

Padi sebagai bahan makanan pokok utama bagi masyarakat Indonesia merupakan tanaman pangan yang rentan terhadap perubahan iklim. Pendataan dan perhitungan ramalan hasil produksi padi sangat diperlukan untuk mendukung kebijakan yang berkaitan dengan ketahanan pangan. Penelitian ini bertujuan untuk melakukan peramalan terhadap produksi padi di Kabupaten Barito Kuala sebagai kabupaten penghasil padi terbesar di Kalimantan Selatan dengan menggunakan data iklim sebagai input. Data iklim yang digunakan berasal dari Stasiun Meteorologi Syamsudin Noor, sedangkan sebagai data output adalah data produksi padi dari Badan Pusat Statistika (BPS) Provinsi Kalimantan Selatan. Metode yang digunakan untuk melakukan peramalan produksi padi adalah Generalized Regression Neural Networks (GRNN). Dari hasil pengujian didapatkan nilai Root Mean Square Error (RMSE) sebesar 0,296 dengan menggunakan parameter smoothness bernilai 1.Kata kunci: padi, iklim, Barito Kuala, GRNN, RMSE

scholarly journals Hydrochars as Emerging Biofuels: Recent Advances and Application of Artificial Neural Networks for the Prediction of Heating Values

Energies ◽  
2020 ◽  
Vol 13 (17) ◽  
pp. 4572
Author(s):  
Ioannis O. Vardiambasis ◽  
Theodoros N. Kapetanakis ◽  
Christos D. Nikolopoulos ◽  
Trinh Kieu Trang ◽  
Toshiki Tsubota ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Sewage Sludge ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Food Waste ◽  
Hydrothermal Carbonization ◽  
Mean Square ◽  
Artificial Neural

In this study, the growing scientific field of alternative biofuels was examined, with respect to hydrochars produced from renewable biomasses. Hydrochars are the solid products of hydrothermal carbonization (HTC) and their properties depend on the initial biomass and the temperature and duration of treatment. The basic (Scopus) and advanced (Citespace) analysis of literature showed that this is a dynamic research area, with several sub-fields of intense activity. The focus of researchers on sewage sludge and food waste as hydrochar precursors was highlighted and reviewed. It was established that hydrochars have improved behavior as fuels compared to these feedstocks. Food waste can be particularly useful in co-hydrothermal carbonization with ash-rich materials. In the case of sewage sludge, simultaneous P recovery from the HTC wastewater may add more value to the process. For both feedstocks, results from large-scale HTC are practically non-existent. Following the review, related data from the years 2014–2020 were retrieved and fitted into four different artificial neural networks (ANNs). Based on the elemental content, HTC temperature and time (as inputs), the higher heating values (HHVs) and yields (as outputs) could be successfully predicted, regardless of original biomass used for hydrochar production. ANN3 (based on C, O, H content, and HTC temperature) showed the optimum HHV predicting performance (R2 0.917, root mean square error 1.124), however, hydrochars’ HHVs could also be satisfactorily predicted by the C content alone (ANN1, R2 0.897, root mean square error 1.289).

scholarly journals Comparison of neural networks and genetic algorithms to determine missing precipitation data (Case study: the city of Sari)

2020 ◽  
Vol 11 (29) ◽  
pp. 114-128
Author(s):  
Ali Mahdavi ◽  
Mohsen Najarchi ◽  
Emadoddin Hazaveie ◽  
Seyed Mohammad Mirhosayni Hazave ◽  
Seyed Mohammad Mahdai Najafizadeh
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Genetic Programming ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Correlation Coefficients ◽  
Mean Square ◽  
Artificial Neural ◽  
The City

Neural networks and genetic programming in the investigation of new methods for predicting rainfall in the catchment area of the city of Sari. Various methods are used for prediction, such as the time series model, artificial neural networks, fuzzy logic, fuzzy Nero, and genetic programming. Results based on statistical indicators of root mean square error and correlation coefficient were studied. The results of the optimal model of genetic programming were compared, the correlation coefficients and the root mean square error 0.973 and 0.034 respectively for training, and 0.964 and 0.057 respectively for the optimal neural network model. Genetic programming has been more accurate than artificial neural networks and is recommended as a good way to accurately predict.

A Study on Time Series Forecasting using Hybridization of Time Series Models and Neural Networks

2020 ◽  
Vol 13 (5) ◽  
pp. 827-832
Author(s):  
Iflah Aijaz ◽  
Parul Agarwal
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Time Series ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Hybrid Models ◽  
Time Series Forecasting ◽  
Percentage Error ◽  
Mean Square

Introduction: Auto-Regressive Integrated Moving Average (ARIMA) and Artificial Neural Networks (ANN) are leading linear and non-linear models in Machine learning respectively for time series forecasting. Objective: This survey paper presents a review of recent advances in the area of Machine Learning techniques and artificial intelligence used for forecasting different events. Methods: This paper presents an extensive survey of work done in the field of Machine Learning where hybrid models for are compared to the basic models for forecasting on the basis of error parameters like Mean Absolute Deviation (MAD), Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Normalized Root Mean Square Error (NRMSE). Results: Table 1 summarizes important papers discussed in this paper on the basis of some parameters which explain the efficiency of hybrid models or when the model is used in isolation. Conclusion: The hybrid model has realized accurate results as compared when the models were used in isolation yet some research papers argue that hybrids cannot always outperform individual models.

Prediction of Significant Wave Height Using Neural Network in the Java Sea (North of Surabaya)

2017 ◽  
Vol 862 ◽  
pp. 72-77
Author(s):  
Wimala L. Dhanistha ◽  
R.A. Atmoko ◽  
P. Juniarko ◽  
Ridho Akbar
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Root Mean Square Error ◽  
Mean Square Error ◽  
Root Mean Square ◽  
Wave Height ◽  
Mean Square ◽  
The North ◽  
Java Sea ◽  
The City

Indonesia is an archipelago, Surabaya is the second crowded city in Indonesia. So the shipping lane and the city is comparable. Neural network is models inspired by biological neural networks and used to estimate or approximate functions that can depend on a large number of inputs and are generally unknown. Neural network is used to predict the wave height in Java Sea (The North of Surabaya). The Root Mean Square Error average for the next one hour is 0.03 and the Root Mean Square Error average for the next six hours is 0.09. That’s mean the longest the prediction, the biggest Root Mean Square error.

scholarly journals LANDSLIDE HAZARD MAPPING USING A RADIAL BASIS FUNCTION NEURAL NETWORK MODEL: A CASE STUDY IN SEMIROM, ISFAHAN, IRAN

2019 ◽  
Vol XLII-4/W18 ◽  
pp. 1085-1090 ◽  
Author(s):  
H. Yavari ◽  
P. Pahlavani ◽  
B. Bigdeli
Keyword(s):  
Neural Network ◽  
Root Mean Square Error ◽  
Radial Basis Function ◽  
Mean Square Error ◽  
Roc Curve ◽  
Basis Function ◽  
Rbf Neural Network ◽  
Landslide Hazard ◽  
Hazard Mapping ◽  
Mean Square

Abstract. In this paper, Radial Basis Function (RBF) Neural Network and Logistic Regression (LR) models were proposed for hazard prediction of landslides in a part of the Semirom area (Iran) to compare their accuracy and performance. For this purpose, a spatial database of the study area was prepared that consists of 68 landslide locations and 11 influencing information layers including slope, aspect, profile curvature, plan curvature, distance from faults, distance from roads, distance from residential regions, distance from rivers, land use, lithology and rainfall. Landslide hazard maps were prepared for the study area by applying the proposed algorithms. Performance of the models was assessed using the Receiver Operating Characteristic (ROC) curve and area under the ROC curve (AUC). The coefficient of determination (R2), the root mean square error (RMSE), and the Normal Root Mean Square Error (NRMSE) were calculated for proposed methods. The outcomes showed that the RBF Neural Network has the highest R2 (0.8224), in comparison to that of the LR model (0.5365). Also, the ROC plots, RMSEs and NRMSEs showed that the proposed RBF Neural Network is much better than the LR model. Consequently, it can be concluded that the RBF Neural Network is the best regression model in this study and it can be considered as a capable method for landslide hazard mapping in landslide-susceptible areas.

Estimation of ambiguous blast-induced ground vibration using intelligent models: A case study

2018 ◽  
Vol 49 (4) ◽  
pp. 147-157 ◽  
Author(s):  
Ragam Prashanth ◽  
DS Nimaje
Keyword(s):  
Neural Network ◽  
Root Mean Square Error ◽  
Radial Basis Function ◽  
Mean Square Error ◽  
Basis Function ◽  
Back Propagation ◽  
Ground Vibration ◽  
Back Propagation Neural Network ◽  
Mean Square ◽  
Intelligent Models

Blasting is an economical and viable operation for reliable excavation of hard rock in mining and civil construction. An ambiguous ground vibration generated by blasting is unenviable and causes grievous damage to nearby inhabitants, residential premises, and other sensitive sites. Accordingly, the proper assessment of indistinct blast-induced ground vibration is a requisite to pinpoint the safe limits in and around mines. An endeavor has been made in this article to apply four predictive models, namely, support vector machine, feed forward back propagation neural network, cascaded forward back propagation neural network, and radial basis function neural network to estimate the ground vibration caused by blasting operation conducted at Mine-A, India. In this article, a total number of 121 blasting operations with relevant parameters are recorded. The most influential parameters of ground vibration are the number of holes, burden, spacing, hole diameter, hole depth, top stemming, maximum explosive charge per delay, and the distance from blast source, which were considered as input parameters. Ground vibration is measured in terms of peak particle velocity and is considered as output. The performance indicators of constructed network models were chosen as the coefficient of determination (R2), root mean square error, and variance account for. Among all constructed intelligent models, the radial basis function neural network with architecture 8-80-1 and R2 of 0.9918, root mean square error of 4.4076, and variance account for of 99.1800 was found to be optimum. Sensitivity analysis showed that the number of holes, burden, and top stemming are the most effective parameters leads to ground vibration due to blasting.

scholarly journals Ανάπτυξη μεθόδων ασαφούς συσταδοποίησης για τη μοντελοποίηση νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης

2011 ◽  
Author(s):  
Αντώνιος Νείρος
Keyword(s):  
Neural Networks ◽  
Radial Basis Function ◽  
Mean Square Error ◽  
Test Data ◽  
Fuzzy Clustering ◽  
Basis Function ◽  
Training Data ◽  
Mean Square ◽  
Information Granulation ◽  
Metric Distance

Στην παρούσα διδακτορική διατριβή λαμβάνει χώρα ερευνητική προσπάθεια στην εκπαίδευση (training) Νευρωνικών Δικτύων και συγκεκριμένα των Νευρωνικών Δικτύων Συναρτήσεων Ακτινικής Βάσης (Radial Basis Function Neural Networks – RBFNN). Για την εκπαίδευση των Νευρωνικών Δικτύων Συναρτήσεων Ακτινικής Βάσης χρησιμοποιήθηκαν αλγόριθμοι ασαφούς συσταδοποίησης (fuzzy clustering). Αναπτύχθηκαν πέντε νέες καινοτόμοι μέθοδοι εκπαίδευσης νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης και οι οποίες έχουν δημοσιευθεί στα πρακτικά διεθνών επιστημονικών συνεδρίων καθώς και σε έγκυρα διεθνή περιοδικά. Η συνεισφορά της παρούσης διδακτορικής διατριβής είναι ότι διεξήχθη συστηματική έρευνα με σκοπό την διερεύνηση της χρήσης της ασαφούς συσταδοποίησης στην εκπαίδευση νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης. Στο πλαίσιο της έρευνας εστιάσαμε στην ανάλυση των πλεονεκτημάτων και των μειονεκτημάτων καθώς και της ουσιαστικής επίδρασης της ασαφούς συσταδοποίησης στην διαδικασία εκπαίδευσης τέτοιου τύπου νευρωνικών δικτύων. Το αποτέλεσμα της εν’ λόγω έρευνας συνίσταται στην ανάπτυξη σε πρώτο επίπεδο τριών απλών μεθόδων ασαφούς συσταδοποίησης για την εκπαίδευση νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης. Συγκεκριμένα στην πρώτη μέθοδο, χρησιμοποιήθηκε η πολύ γνωστή μέθοδος των Ασαφών c-Μέσων (FCM) για την προεπεξεργασία των δεδομένων και στη συνέχεια η μέθοδος των σταθμισμένων (weighted) Ασαφών c-Μέσων (FCM) για τον υπολογισμό των παραμέτρων των πυρήνων των κρυφών κόμβων του νευρωνικού δικτύου. Στη δεύτερη μέθοδο, χρησιμοποιήθηκε η βέλτιστη συσταδοποίηση (optimal clustering) για τον καθορισμό του αριθμού των πυρήνων των συναρτήσεων ακτινικής βάσης καθώς και τον παραμέτρων του νευρωνικού δικτύου. Τέλος προτάθηκε μια τρίτη μέθοδος που συνδυάζει ασαφή συσταδοποίηση και τη βέλτιστη ασαφή συσταδοποίηση για τον αποτελεσματικό σχεδιασμό νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης. Σε δεύτερο επίπεδο, αναπτύχθηκαν δύο καινοτόμες μέθοδοι εκπαίδευσης νευρωνικών δικτύων συναρτήσεων ακτινικής βάσης. Συγκεκριμένα στην πρώτη μέθοδο προτείνεται μια νέα καινοτόμος υβριδική μέθοδος συσταδοποίησης, η οποία συνδυάζει με ομοιόμορφο τρόπο την διακριτή (crisp) και την ασαφή (fuzzy) συσταδοποίηση (clustering) για τον υπολογισμό των παραμέτρων των πυρήνων των συναρτήσεων ακτινικής βάσης του νευρωνικού δικτύου. Η δεύτερη προτεινόμενη μέθοδος χρησιμοποιεί κοκκοποίηση πληροφορίας (information granulation) για τον υπολογισμό των κέντρων των πυρήνων των συναρτήσεων ακτινικής βάσης και μία νέα μετρική απόσταση (metric distance) για τον υπολογισμό των πλατών των συναρτήσεων ακτινικής βάσης. Σε όλες τις προτεινόμενες μεθόδους έγινε πρακτική εφαρμογή σε πραγματικά μοντέλα και σε προσεγγίσεις συναρτήσεων, τα αποτελέσματα των οποίων έδειξαν ότι οι προτεινόμενες μέθοδοι παράγουν αξιόπιστα, ακριβή και συμπαγή νευρωνικά δίκτυα συναρτήσεων ακτινικής βάσης. Η σύγκριση με την διεθνή βιβλιογραφία έδειξε ότι τα δίκτυα RBF που δημιουργήθηκαν έχουν καλύτερα αποτελέσματα όσον αφορά το μέσο τετραγωνικό σφάλμα (mean square error – MSE) καθώς και το κανονικοποιημένο μέσο τετραγωνικό σφάλμα (normalized mean square error – NRMSE) τόσο για τα δεδομένα εκπαίδευσης (training data) όσο για τα δεδομένα δοκιμής (test data).

Sign in / Sign up

Export Citation Format

Share Document