This paper provides a comprehensive survey on concept drift adaptation, a critical issue in online supervised learning where the relationship between input data and the target variable changes over time. The authors, from various institutions, introduce the concept of concept drift, define adaptive learning algorithms, and discuss their application in different domains. They categorize existing strategies for handling concept drift, present representative techniques and algorithms, and evaluate their performance. The survey aims to unify the terminology and concepts related to concept drift adaptation, covering different facets of the problem to reflect the current state of the art. The paper is organized into sections that cover the problem setting, adaptive learning algorithms, experimental settings, and illustrative applications. It also includes a taxonomy of methods for concept drift adaptation, focusing on memory, change detection, learning, and loss estimation. The authors highlight the importance of understanding the nature and source of drift to engineer effective adaptive learning strategies.This paper provides a comprehensive survey on concept drift adaptation, a critical issue in online supervised learning where the relationship between input data and the target variable changes over time. The authors, from various institutions, introduce the concept of concept drift, define adaptive learning algorithms, and discuss their application in different domains. They categorize existing strategies for handling concept drift, present representative techniques and algorithms, and evaluate their performance. The survey aims to unify the terminology and concepts related to concept drift adaptation, covering different facets of the problem to reflect the current state of the art. The paper is organized into sections that cover the problem setting, adaptive learning algorithms, experimental settings, and illustrative applications. It also includes a taxonomy of methods for concept drift adaptation, focusing on memory, change detection, learning, and loss estimation. The authors highlight the importance of understanding the nature and source of drift to engineer effective adaptive learning strategies.