AI-Powered Data Imputation Techniques for Handling Missing Values in Large Enterprise Datasets

Abstract

Handling missing values in large enterprise datasets is a critical challenge in data science, as incomplete data can lead to biased analyses, inaccurate predictions, and compromised decision-making. Traditional imputation techniques, such as mean substitution and regression-based methods, often fail to capture underlying data distributions. AI-powered data imputation techniques, including deep learning-based methods, probabilistic models, and generative adversarial networks (GANs), provide a more accurate and scalable solution. This paper explores recent advancements in AI-driven data imputation, compares various techniques, and evaluates their performance in enterprise environments.