Amazing technological breakthrough possible @S-Logix

Office Address

  • #5, First Floor, 4th Street Dr. Subbarayan Nagar Kodambakkam, Chennai-600 024 Landmark : Samiyar Madam
  • +91- 81240 01111

Social List

Research Topics in Deep Learning for Intrusion Detection System

Research Topics in Deep Learning for Intrusion Detection System

  Intrusion Detection Systems (IDS) are very important network security tools to remove threats that would otherwise occur when carrying information, prevent unauthorized access or abuse, and report attacks to those responsible for security. The main goal of Intrusion Detection Systems is to detect and classify intrusions, attacks, or violations of the security policies automatically at network-level and host-level infrastructure promptly. IDS plays a crucial role in the security of the networks, consisting of three main components: data collection, feature selection/conversion, and decision engine. Classification of IDSs, according to their techniques- Signature-Based and Anomaly-Based and according to their location- Host-based and network-based.

  Deep learning is an effective approach in intrusion detection due to less training time and high accuracy. The deep learning model learns the abstract and high dimensional feature representation of the IDS data by passing them into many hidden layers. A deep neural network (DNN) is explored to develop a flexible and effective IDS to detect and classify unforeseen and unpredictable cyber-attacks. Convolutional Neural Network, Long Short Term Memory Neural Network, Autoencoders, and Deep belief network are the mainly used deep algorithms for classifying the intrusion. Future Advancements in IDS using deep learning are IDS on big data to explore low-quality data, IDS based on self-learning deep learning models, IDS on deep learning models with generalization, among others.

  • Deep learningĀ is vastly employed by intrusion detection systems (IDS) to improve the security of computer networks and hosts.

  • A deep learning-based intrusion detection system involves two major tasks such as extraction of features and classification tasks.

  • Deep learning allows deep neural networks (DNNs) facilitate to develop an effective IDS with the learning capability to detect recognized and new or zero-day network behavioral features, consequently ejecting the systems intruder and reducing the risk of compromise.

  • Deep learning methods determine the relevant features among the data using feature selection and extraction and automatically discover the essential differences between normal and abnormal data with high accuracy.

  • Due to the dynamic nature of malware with continuously changing attacking methods, Deep learning can cope with large-scale data and has shown success in various ways.