The remarkable ability to separate and identify self and non-self in a given problem space, makes negative selection a fascinating concept of artificial immune system. Therefore, negative selection has attracted research interest and is studied and explored for complex problem solving across different application areas. Anomaly detection in computer security is a thriving area of research and has witnessed various new explores involving different computational intelligence techniques. Negative selection with its core ability to detect self and non-self along with traits like adaptability, learning, robustness and faster response makes it a suitable and compelling concept for anomaly detection. Over the years, negative selection has evolved from its preliminary theories and it has embraced new improvements from computational intelligence in its several concepts. This paper intends to review various negative selection taxonomies, representations and matching techniques from inception to current scenario in anomaly detection. It attempts to critically evaluate and classify available literature to establish future areas of research for formulating potential solutions to mitigate the complex security challenges.

在给定的问题空间中分离和识别自我和非自我的非凡能力，使否定选择成为人工免疫系统的一个迷人概念。因此，否定选择引起了研究兴趣，并被研究和探索用于解决不同应用领域的复杂问题。计算机安全中的异常检测是一个蓬勃发展的研究领域，见证了涉及不同计算智能技术的各种新探索。负选择具有检测自我和非自我的核心能力以及适应性、学习性、稳健性和更快的响应等特征，使其成为异常检测的合适且引人注目的概念。多年来，负选择已从其初步理论发展而来，并在其多个概念中接受了计算智能的新改进。本文旨在回顾异常检测中从开始到当前场景的各种负选择分类、表示和匹配技术。