Detection of Network Security Traffic Anomalies Based on Machine Learning KNN Method
Fanyi Zhao, Mingxuan Zhang, Shiji Zhou, Qi Lou
Abstract
This paper discusses the application and advantages of machine learning in anomaly detection of network security traffic. By summarizing the existing methods and techniques of network anomaly detection, this paper focuses on the progress of clustering, classification, statistics, and information theory in research. In particular, innovations in data preprocessing, feature selection, and algorithm design, such as experimental validation based on an improved KNN algorithm, demonstrate the potential of machine learning in improving detection accuracy and efficiency. In the future, as the amount of data increases and algorithms are further optimized, these technologies are expected to drive further development in cybersecurity and address the challenges of increasingly complex cyber threats.