Ethical Frameworks for Information Systems: Integrating Social Science Principles with Computational Models

Abstract

This study presents a view on how ethical frameworks can be facilitated through artificial intelligence (AI) systems, with a specific focus on responsible AI practices and healthcare, finance, and government application. The research explores the performance of Decision Trees, Neural Networks, SVM, as well as Random Forest, in relation to ethical decision-making processes. The experimental data unveils that Random Forest was the most accurate at 94%, followed by Neural Networks at 91%, SVM at 87%, and Decision Trees at 83%. The evaluation favored fairness, transparency and elimination of bias, with an immense focus on preventing harm and outcomes equivalent in value. The results show that there is need for regulatory standards and ethical protocols in ensuring openness and accountability with the implementation of AI technologies. It is obvious that ensemble methods particularly Random Forest are most proficient in a complex deal of ethical questions. In real life, Decision Trees could be helpful, but quite ineffective when it comes to cases with high consequences. In order to contribute to developing AI-based technologies which have positive effects on the society and avert unpredictable threats, this research explores the way AI supports ethical principles.