Application of Multivariate Adaptive Regression Splines and Classification and Regression Trees to Estimate Wave-Induced Scour Depth Around Pile Groups

Samadi, Mehrshad; Afshar, M. H.; Jabbari, Ebrahim; Sarkardeh, Hamed

doi:10.1007/s40996-020-00364-2

Cited by 21 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multivariate Adaptive Regression Spline (MAR) is a versatile regression method and nonparametric approach that incorporates piecewise linear regression function referred to as basic function (bf). To estimate the performance of MAR, it uses basic functions (bf) for capturing the hidden nonlinear relations between independent input variables [37]. Bf is therefore the main component in the generation of a MAR model.…”

Section: Multivariate Adaptive Regression Spline (Mar)mentioning

confidence: 99%

Scientometric Analysis of COVID-19 Scholars Publication using Machine Learning

Oyewola¹,

Dada

2022

International Journal of Applied Mathematics Electronics and Computers

View full text Add to dashboard Cite

The global health crisis that started in December 2019 resulted in an outbreak of coronavirus named COVID-19. Scientists worldwide are working to demystify the transmission and pathogenic mechanisms of the deadly coronavirus. The World Health Organization has declared COVID-19 a pandemic in March 2020, which makes it essential to track and analyse the research state of COVID-19 for guidance on further research. This research was conducted using scientometric analysis, knowledge-mapping analysis, COVID-19 studies and journal classifications. The publications used in this study include over 3000 COVID-19 papers made available to the public from 1 January 2018 to 15 April 2021 in the PubMed databases. In this study, it was discovered that the rapid reaction of researchers worldwide resulted in a fast growth trend between 2019 and 2021 in the number of publications related to COVID-19. It was discovered that the largest number of studies is in the United States of America, which is one of the countries most affected by a pandemic. The method adopted for this study involved the use of documents such as Case Reports (CAT), Journal Article (JAT), letter (LTR), EAT, and Editorial (EDT). This is followed by the classification of COVID-19 related publications that were retrieved from PubMed between 2019 and 2021 using machine learning (ML) models such as Naïve Bayes (NB), Bayesian Generalized Linear Model (BGL), Heteroscedastic Discriminant Analysis (HDA) and Multivariate Adaptive Regression Spline (MAR). Simulation results show that the classification accuracy of MAR is better than that of other ML models used in this study. The sensitivity of the MAR is within the range of 100%. This shows that MAR performs better than NB, BGL and HDA. MAR performs better with an overall accuracy of 89.62%. Our results show a high degree of strong collaboration in coronavirus research and the exchange of knowledge in the global scientific community.

show abstract

Section: Multivariate Adaptive Regression Spline (Mar)mentioning

confidence: 99%

Scientometric Analysis of COVID-19 Scholars Publication using Machine Learning

Oyewola¹,

Dada

2022

International Journal of Applied Mathematics Electronics and Computers

View full text Add to dashboard Cite

show abstract

“…In complicated engineering problems, data-driven methods, such as artificial neural networks (ANNs); adaptive neurofuzzy inference system (ANFIS); wavelet-hybrid (W-hybrid) data-driven methods; evolutionary polynomial regression (EPR); Support Vector Machines (SVMs); classification and regression trees (CART); multivariate adaptive regression splines (MARS); gene expression programming (GEP) and group method of data handling (GMDH), which are frequently used and have been applied in many civil engineering fields, especially in water-related problems (Samadi et al 2015(Samadi et al , 2020a(Samadi et al , 2020bMojaradi et al 2018;Torabi et al 2022).…”

Section: Introductionmentioning

confidence: 99%

“…Generally, the data-driven models are classified into white-box and black-box models. White-box data-driven models give mathematical equations that enable the relationship between the variables and dependent variables to be understood and interpreted directly (Samadi et al 2020a(Samadi et al , 2020b. By contrast, black-box models predict output parameters using numerical values rather than a straightforward equation between the input and output variables (Samadi et al 2021a(Samadi et al , 2021b.…”

Section: Introductionmentioning

confidence: 99%

Estimating the permeability coefficient of soil using CART and GMDH approaches

2022

Self Cite

View full text Add to dashboard Cite

Permeability coefficient of soil (k) is one of the most important parameters in groundwater studies. This study, two robust explicit data-driven methods, including Classification and Regression Trees (CART) and the Group Method of Data Handling (GMDH) were developed using characteristics of soil, i.e., clay content (CC), water content (ω), liquid limit (LL), plastic limit (PL), specific density (γ), void ratio (e) to generate predictive equations for prediction of k. When compared to CART; Mean Absolute Error (MAE) = 0.0051, Root Mean Square Error (RMSE) = 0.0088, Scatter Index (SI) = 64.00%, Correlation Coefficient (R) = 0.7841, Index agreement (Ia) = 0.8830; the GMDH equation produced the lowest error values; MAE = 0.0044, RMSE = 0.0072, SI = 52.17%, R = 0.8493, Ia = 0.9184; in testing stage. Although, GMDH had better performance, however, CART and GMDH could be considered effective approaches for the prediction of k.

show abstract

“…The main objective of this research is to investigate the utilization of third‐order morphometric variables and multi‐scale geomorphometric algorithms in soil thickness modeling based on morphometric variables using multivariate adaptive regression splines (MARS). MARS is a relatively novel data‐driven technique that considers the interaction and nonlinear relationships between independent and dependent variables and produces a flexible and comprehensible model for prediction by combining simple linear regression equations (Samadi, Afshar, Jabbari, & Sarkardeh, 2020). The successful application of MARS to solving different problems has been reported by some researchers (Emamgolizadeh, Bateni, Shahsavani, Ashrafia, & Ghorbania, 2015; Muñoz & Felicísimo, 2004; Samadi et al., 2020).…”

Section: Introductionmentioning

confidence: 99%

“…MARS is a relatively novel data‐driven technique that considers the interaction and nonlinear relationships between independent and dependent variables and produces a flexible and comprehensible model for prediction by combining simple linear regression equations (Samadi, Afshar, Jabbari, & Sarkardeh, 2020). The successful application of MARS to solving different problems has been reported by some researchers (Emamgolizadeh, Bateni, Shahsavani, Ashrafia, & Ghorbania, 2015; Muñoz & Felicísimo, 2004; Samadi et al., 2020). However, MARS has been used in soil–landscape modeling less than other data‐driven methods.…”

Section: Introductionmentioning

confidence: 99%

Utilization of multi‐scale geomorphometric algorithms and third‐order morphometric variables for soil thickness modeling

Khanifar

Khademalrasoul

2022

Transactions in GIS

View full text Add to dashboard Cite

The significance of multi‐scale geomorphometric algorithms and third‐order morphometric variables was studied in soil thickness modeling. In this research, the algorithms of Evans–Young, Shary, Zevenbergen–Thorne, and Florinsky were employed to calculate morphometric variables on different neighborhood scales. The results demonstrate that the polynomial model and neighborhood scale can have a remarkable effect on morphometric variables. Based on correlation analysis, we observed that on a given neighborhood scale, not every polynomial model might be appropriate for modeling soil thickness. A geomorphometric algorithm that provides a modest smoothing rate of morphometric variables can be ideal for soil–landscape modeling. The results revealed that Gnn, a third‐order morphometric variable, was the most relevant predictor of soil thickness. The third‐order morphometric variables have the potential to be considered for soil–landscape modeling. Multivariate adaptive regression splines provided a model (R2 = 0.71, RMSE = 0.28 m) for predicting soil thickness based on local morphometric variables that were calculated using two types of polynomial models. Based on the findings of this study, it is recommended that several polynomial models and neighborhood sizes be employed when performing soil–landscape analysis. Consideration of multi‐scale geomorphometric analysis can increase the performance of digital soil modeling.

show abstract

Application of Multivariate Adaptive Regression Splines and Classification and Regression Trees to Estimate Wave-Induced Scour Depth Around Pile Groups

Cited by 21 publications

References 46 publications

Scientometric Analysis of COVID-19 Scholars Publication using Machine Learning

Scientometric Analysis of COVID-19 Scholars Publication using Machine Learning

Estimating the permeability coefficient of soil using CART and GMDH approaches

Utilization of multi‐scale geomorphometric algorithms and third‐order morphometric variables for soil thickness modeling

Contact Info

Product

Resources

About