A Proposed Method for Detecting Network Intrusion Using Deep Learning Approach

NIDSs, known as network intrusion detection systems, are essential for protecting computer networks. Nonetheless, there are concerns about the sustainability and viability of current approaches for meeting the needs of modern networks. These issues are more specifically connected to the decreased detection accuracy and the increased human involvement needed. This paper introduces a new deep-learning intrusion detection method to address these problems. We use a deep learning method by creating a Deep Neural Network (DNN) model for detecting intrusions and training it using the NSLKDD Dataset. From the 41 features in the NSL-KDD Dataset, we only use 37 of the essential features in this work. We demonstrate from various studies that the deep learning approach has much potential for use in NIDs. In this paper, we show the efficiency of our method and compare it with previous studies in terms of precision, accuracy, recall, and f-measure values.