User:Aydinshishegaranam/sandbox

=Aydin Shishegaran=

He is a Ph.D. candidate in environmental engineering and civil engineering at Iran University of Science and Technology. He has published more than 11 papers in structural and environmental engineering. He has used machine learning (ML), finite element method (FEM), multi-criteria decision analysis (MCDA) in his research. .

In 2020, he introduced a new machine learning method titled "High correlated variable creator machine". This algorithm is a novel hybrid model to improve regression models and artificial neural network models to predict phenomena and material performance. HCVCM tries to create new variables instead of the initial variables, which are more effective in improving the accuracy of models. It generates new variables from the initial variables using mathematical functions, such that they have more correlation with the output and less correlation with other inputs. There are three steps in HCVCM. First, several mathematical functions create new variables. Secondly, the new variables are selected, such that they have more correlation with the output in comparison to the initial variables, and they are imported in the third step, in which only the new variables are imported to the next generation or the regression model. Their correlation with other inputs is less than the correlation between the initial variables.

Moreover, he introduced a new multi-criteria decision analysis (MCDA) method entitled "applied effect of changes intensity in each indicator (AECIEI)" in 2020. In this method, not only the effect of the weight of indicators is used on sustainability evaluation, but the effect of intensity of variation of indicators is applied to rank the scenarios. .