Intrusion Detection in Big Data Environment. PROFICIENT INTRUSION DETECTION SYSTEM USING MAP REDUCE BASED DEEP LEARNING IN BIG DATA ENVIRONMENT

      Аннотация: Nowadays, information system management, large-scale data-clustering and classification have become increasingly important and a challenging area. While several tools and methods have been proposed, few of them are sufficient and efficient enough for real applications due to the rapid growing-in-size and high-dimensional data inputs. Intrusion in lay terms is unwanted or unauthorized hacker interference, and as it is unwanted or unauthorized, and mostly with bad intentions such they may misuse the resources. The intrusion intends to collect information related to the company, such as the structure of the internal networks or software systems such as operating systems, tools/utilities, or software applications used by the corporation and then initiate connections to the internal network and carry out attacks. Intrusions are carried out by people who are not related to the company or the business’s external. Intrusion detection is a process of observing the activities from the computer-based system and scrutinizing them for possible signs of incidents which are violations of security policies, guidelines, or standard security practices.

Ключевые слова: Data Mining, Intrusion Detection, Recurrent Neural Network, Support Vector Machine, Mutual Information, Mean squared Error, Machine Learning, extreme learning machine