Application of Machine Learning algorithms for experimental data processing and estimation of 96Mo(n, p)96Nb reaction cross section

Abstract

In this paper, Machine learning techniques have been employed for preparation and estimation of 96 Mo (n, p) 96Nb reaction data. The experimental data of 96 Mo (n, p) 96Nb reaction available in the EXFOR database was retrieved, analyzed and processed using renormalization and data cleaning techniques. Estimation of the renormalized experimental data with outlier and without outlier, over the entire neutron energy range, was then performed using machine learning regression algorithms of Ordinary Least square, Ridge, Least Absolute Shrinkage and Selection Operator (LASSO) and Support Vector Regressor. The results obtained were then compared and it was observed that the data preparation plays a significant role in data quality.