Artificial Intelligence and Machine Learning have gained attention due to their increasing applications in various areas of life. However, many of these models are difficult for humans to interpret, posing problems in critical situations such as loan approvals, surgeries, and legal decisions. Explainable AI (XAI) aims to address these concerns by providing human-interpretable explanations. We aim to explore the use of XAI techniques beyond their original purpose, specifically investigating if they can offer insights into datasets themselves. This would support existing data analysis methods like regression analysis, commonly used to estimate relationships between variables.
1. “Explainable Artificial Intelligence as a Generic Data Analysis Tool” at IC2S2 2023, Copenhagen, IL