NONLINEAR REGRESSION MODEL FOR ESTIMATING THE SIZE OF WEB-APPLICATIONS CREATED USING THE LARAVEL FRAMEWORK
Sergiy B. Prykhodko, Natalia Prykhodko, Mykhaylo V. Vorona, I. A. Belovol · Information Technology and Computer Engineering · 2021
The three-factor nonlinear regression model to estimate the size of development of web applications created using the Laravel framework, is constructed on the basis of normalization of the four-dimensional non-Gaussian data set (actual size in KLOC; number of classes, sum of average afferent coupling and average efferent coupling; average number of methods) by the Johnson multivariate transformation for SB family. Comparison of the constructed model with the linear regression model and nonlinear regression models based on the decimal logarithm and the Johnson univariate transformation is performed. Thе constructed model, in comparison with other regression models, has a smaller value of the mean magnitude of the relative error and smaller widths of the prediction intervals of nonlinear regression.