Pipeline Tiền Xử Lý Dữ Liệu Mạng Cho Mô Hình CNN Phát Hiện Xâm Nhập Đa Lớp

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Applying deep learning technology to build models to detect cyberattacks is a popular trend. In particular, visualising network data to create images familiar to CNN (Convolutional Neural Network) to have good classification as in computer vision is a very creative way. However, detecting many types of attacks using multilayer CNN classification models with imaged network datasets is still limited. In this paper, we propose a method to preprocess network data before converting it into images for input into multilayer CNN deep learning networks. This method was applied to process the NSL-KDD dataset with the separate KDDTest+ and showed competitive results. This result also shows that data preprocessing is still a necessary step when applying deep learning to building a network attack detection model.