Support vector machines for modelling phosphocalcic hydroxyapatite by precipitation from a calcium carbonate solution and phosphoric acid solution

Abstract

The aim of this work was to develop an effective quantitative relationship model using support vector machines for the synthesis of phosphocalcic hydroxyapatite by precipitation from a calcium carbonate solution and a phosphoric acid solution. The model was created using a set of factors consisting of the pH of the reaction solution, the Ca/P molar ratio of the reagents, the reaction time and the initial concentration of calcium. Previous studies modelled these relationships using different classification techniques. In this paper, a novel approach based on support vector machines (SVMs) is introduced. This latter approach yields the best result compared to polynomial regression (PR), linear regression (LR) and artificial neural networks (ANN). In addition, the contribution of each descriptor is evaluated. Thus, the proposed method can be successfully used to predict the Ca/P analysis directly from the considered factors.

Development of Crack Width Prediction Models for RC Beam-Column Joint Subjected to Lateral Cyclic Loading Using Machine Learning

“…In other words, it shows how clustered the data is around the best fit axis, thus supporting the argument through a previous study in which RMSE was able to avoid over-fitting in positive identification [62,65]. This parameter of the RMSE index is, therefore, a simple, quick, and reliable metric compared to other parameters, widely used in climate analysis, forecasting, and regression to validate experimental results [50,64,94].…”

“…For the SVM model, the compressive strength and prediction of autogenous shrinkage for concrete were developed, respectively, using the regression technique [62,63]. One of the distinguishing features of SVM is its limited number of parameters compared to other types of prediction models that need a design that must integrate network structure and be combined with powerful optimization algorithms to deliver satisfactory results [64]. In these researches, both authors [62,63] applied five and nine numbers of parameters, respectively, to further simulate and correlate the predicted outcome with the experimental results.…”

Machine learning methods in predicting the student academic motivation

“…In the first step, the maximum margin classifier is introduced for problems where a linear decision boundary can be used for distinguishing two classes, and then in second step, the classifier is altered to allow misclassification, while in third step, using the kernel classifier is nonlinearly generalized [15]. As stated in [19], for a given set of data (x1,y1)..(xk,yk), ∈ as input space, ∈ {−1,1} as class labels, the objective is to form a hyperplane…”

“…is a kernel function that performs mapping ∈ . Three different kernel functions were applied in this research: The Radial Basis Function (RBF) kernel [19] K x i ,x j = exp -γ x i -x j 2 , γ is a width parameter (γ > 0) (11) sigmoid kernel [35] K x i ,x j = tanh γx i x j +coefficient , γ>0, coefficient ≥0…”

SYNTHESIS OF ULTRAFINE HYDROXYAPATITE IN A Ca(OH)2-H3PO4-H2O REACTION SYSTEM USING ULTRASOUND IRRADIATION

“…Hydroxyapatite (HAp) is very insoluble, and is used as, e.g., a biomaterial, ion-exchange material and catalyst. In general, HAp is synthesized by solid-phase reaction at around 1000 ºC, but synthesis is possible at 100 ºC or lower by liquid-phase reaction [1][2][3][4] . The phosphate used in the synthesis of HAp is important.…”