Oral Presentations: UC 330
Classifying Abnormal Network Captures with Early Response: Artificial Intelligence Defense Systems
Presentation Type
Presentation
Faculty Mentor’s Full Name
Douglas Raiford
Abstract / Artist's Statement
Society is steadily being confronted with an increasing number of issues which bring the concept of security into question. Cyberwarfare is an important issue for our generation, and as humanity progresses towards the development of programs and algorithms that utilize artificial intelligence, it is imperative to incorporate machine learning algorithms into our intrusion detection systems. Artificial intelligence can be used to minimize the human element, which is often the weakest portion of a system, in many different situations.
This study will test the accuracy and efficiency of using multiple different machine learning algorithms to analyze sets of normal and abnormal network traffic and predict the likeliness that any particular instance of activity on a network is a malicious intruder. I will be comparing the results from a few machine learning algorithms and finding which algorithm has the most efficiency, and then comparing this to the efficiency of human monitoring.
This research will help contribute to the field of Cybersecurity by displaying how artificial intelligence could be used to increase security for nearly any network or system and help propel the combination of the Artificial Intelligence and Cybersecurity fields into the future. I will be comparing my findings to human monitoring.
Category
Humanities
Classifying Abnormal Network Captures with Early Response: Artificial Intelligence Defense Systems
Society is steadily being confronted with an increasing number of issues which bring the concept of security into question. Cyberwarfare is an important issue for our generation, and as humanity progresses towards the development of programs and algorithms that utilize artificial intelligence, it is imperative to incorporate machine learning algorithms into our intrusion detection systems. Artificial intelligence can be used to minimize the human element, which is often the weakest portion of a system, in many different situations.
This study will test the accuracy and efficiency of using multiple different machine learning algorithms to analyze sets of normal and abnormal network traffic and predict the likeliness that any particular instance of activity on a network is a malicious intruder. I will be comparing the results from a few machine learning algorithms and finding which algorithm has the most efficiency, and then comparing this to the efficiency of human monitoring.
This research will help contribute to the field of Cybersecurity by displaying how artificial intelligence could be used to increase security for nearly any network or system and help propel the combination of the Artificial Intelligence and Cybersecurity fields into the future. I will be comparing my findings to human monitoring.