Projects
Multi-Objective Feature Selection in Credit Scoring
Multi-Objective Feature Selection in Credit Scoring was the title of my Master thesis at Humboldt University of Berlin. The thesis describes a new feature selection approach based on a multi-objective genetic optimization algorithm. An implementation of the Multi-Objective NSGA-III algorithm for feature selection in R programming language was presented as a part of master thesis and published at CRAN. This article presents a brief overview of the proposed approach as well as key findings.
Technologies: R Feature_Engineering Genetic_Algorithm XGBoost