Abstract

The reason for this study is the complexity of data analysis models, emphasizing the fundamental importance of explainability, especially in areas driven by Artificial Intelligence (AI). The goal is to examine how reducing bias affects explainability. Additionally, it explores how simplified models can obtain functionally similar results, making them more understandable and acceptable. Integrating bias reduction methods with statistical techniques and decision support systems, it’s intended to improve the acceptance of models by reducing concerns about biased results. The effectiveness of simplification strategies to increase transparency and understanding is analysed through quantitative and qualitative evaluation. The participation of those involved and ethical principles are essential to this method. The research seeks to help improve transparency in data analysis and solve important social problems.

Download

Share

COinS