Scribd
Upload
65 views12 pages

AI Neural Networks for Cyber Threat Detection

This document presents a proposal to develop an AI-based cyber threat detection system using artificial neural networks. It aims to improve upon existing signature-based detection methods by applying deep learning algorithms to analyze large amounts of security event data. Specifically, the proposed system would preprocess event data into profiles, then use neural network models like CNN, LSTM and FCNN to learn normal and threat patterns to detect unknown cyber threats. The authors believe this approach can achieve better accuracy than traditional machine learning methods. They compare different deep learning algorithms and evaluate their performance at detecting malicious attacks.

Uploaded by

Mr.single KANYA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
65 views12 pages

AI Neural Networks for Cyber Threat Detection

This document presents a proposal to develop an AI-based cyber threat detection system using artificial neural networks. It aims to improve upon existing signature-based detection methods by applying deep learning algorithms to analyze large amounts of security event data. Specifically, the proposed system would preprocess event data into profiles, then use neural network models like CNN, LSTM and FCNN to learn normal and threat patterns to detect unknown cyber threats. The authors believe this approach can achieve better accuracy than traditional machine learning methods. They compare different deep learning algorithms and evaluate their performance at detecting malicious attacks.

Uploaded by

Mr.single KANYA
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 12

MALLA REDDY INSTITUTE OF TECHNOLOGY

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

CYBER THREAT DETECTION


BASED ON ARTIFICIAL NEURAL
USING EVENT PROFILES
NETWORKS

PRESENTED BY:
R.Shalem Raj
R.Kanya Naik
R.Manivardhan Reddy
P.Sriram
CONTENTS

 ABSTRACT
 INTRODUCTION

 OBJECTIVE
 EXISTING SYSTEM

 DISADVANTAGES OF EXISTING SYSTEM

 PROPOSED SYSTEM
 ADVANTAGES OF PROPOSED SYSTEM
 SOFTWARE AND HARDWARE REQUIREMENTS
ABSTRACT
• ONE OF THE MAJOR CHALLENGES IN CYBERSECURITY IS THE PROVISION OF
AN AUTOMATED AND EFFECTIVE CYBER-THREATS DETECTION TECHNIQUE.
• SO OVER A NETWORK , BETWEEN A SENDER AND RECEIVER THEIR SHOULD
BE A THREAT DETECTION MECHANISM TO DETECT CYBER THREATS.
• FOR THIS WORK, WE DEVELOPED AN AI-SIEM SYSTEM BASED ON A COMBINATION OF
EVENT PROFILING FOR DATA PRE-PROCESSING AND DIFFERENT ARTIFICIAL NEURAL
NETWORK METHODS, INCLUDING FCNN, CNN, AND LSTM.

• WE ARE COMPARING DEEP LEARNING ALGORITHMS WITH MACHINE


LEARNING ALGORITHMS TO CHECK ACCURACY RATE.
INTRODUCTION
• DUE TO THE MONUMENTAL GROWTH OF INTERNET APPLICATIONS IN
THE LAST DECADE, THE NEED FOR SECURITY OF INFORMATION
NETWORK HAS INCREASED MANIFOLDS.
• AS A PRIMARY DEFENSE OF NETWORK INFRASTRUCTURE, AN
INTRUSION DETECTION SYSTEM IS EXPECTED TO ADAPT TO
DYNAMICALLY CHANGING THREAT LANDSCAPE.
• MANY SUPERVISED AND UNSUPERVISED TECHNIQUES HAVE BEEN
DEVISED BY RESEARCHERS FROM THE DISCIPLINE OF MACHINE
LEARNING.
• DEEP LEARNING IS AN AREA OF MACHINE LEARNING WHICH APPLIES
NEURON-LIKE STRUCTURE FOR LEARNING TASKS.
• DEEP LEARNING HAS PROFOUNDLY CHANGED THE WAY WE APPROACH
LEARNING TASKS BY DELIVERING MONUMENTAL PROGRESS IN
DIFFERENT DISCIPLINES LIKE SPEECH PROCESSING, COMPUTER VISION,
AND NATURAL LANGUAGE PROCESSING AND MANY MORE FIRMS.
• IT IS ONLY RELEVANT THAT THIS NEW TECHNOLOGY MUST BE
INVESTIGATED FOR INFORMATION SECURITY APPLICATIONS.
• SO WE BELIEVE THAT DEEP LEARNING ALGORITHMS USING ARTIFICIAL
NEURAL NETWORKS CAN PERFORM BETTER THAN SUPERVISED
ALGORITHMS OF MACHINE LEARNING ALGORITHMS AND CAN GET
BETTER ACCURACY.
OBJECTIVE
• THE MAIN OBJECTIVE OF THIS PROJECT IS TO PERFORM THE MECHANISM USING
DEEP LEARNING ALGORITHMS .
• ALSO TO GET BETTER ACCURACY COMPARED TO MACHINE LEARNING ALGORITHMS
SO THAT CYBER THREAT CAN BE DETECTED WITH MAJORITY OF ACCURACY.
EXISTING SYSTEM
• TRADITIONALLY, THERE ARE TWO PRIMARY SYSTEMS FOR DETECTING
CYBER-THREATS AND NETWORK INTRUSIONS.
• AN INTRUSION PREVENTION SYSTEM (IPS) IS INSTALLED IN THE ENTERPRISE
NETWORK, AND CAN EXAMINE THE NETWORK PROTOCOLS AND FLOWS WITH
SIGNATURE-BASED METHODS PRIMARILY. IT GENERATES APPROPRIATE
INTRUSION ALERTS, CALLED THE SECURITY EVENTS, AND REPORTS THE
GENERATING ALERTS TO ANOTHER SYSTEM, SUCH AS SIEM.
• THE SECURITY INFORMATION AND EVENT MANAGEMENT (SIEM) HAS BEEN
FOCUSING ON COLLECTING AND MANAGING THE ALERTS OF IPSS.
DISADVANTAGES OF EXISTING
SYSTEM
 IT IS STILL DIFFICULT TO RECOGNIZE AND DETECT INTRUSIONS
AGAINST INTELLIGENT NETWORK ATTACKS OWING TO THEIR HIGH
FALSE ALERTS AND THE HUGE AMOUNT OF SECURITY DATA

 THESE LEARNING-BASED APPROACHES REQUIRE TO LEARN THE


ATTACK MODEL FROM HISTORICAL THREAT DATA AND USE THE
TRAINED MODELS TO DETECT INTRUSIONS FOR UNKNOWN CYBER
THREATS.

 HENCE MACHINE LEARNING ALGORITHMS SOME TIMES DISRUPTS


TO DETECT MALICIOUS ATTACKS
PROPOSED SYSTEM
 WE DEVELOPED A GENERALIZABLE SECURITY EVENT ANALYSIS METHOD
BY LEARNING NORMAL AND THREAT PATTERNS FROM A LARGE AMOUNT OF
COLLECTED DATA, CONSIDERING THE FREQUENCY OF THEIR OCCURRENCE.

 SO THAT DETECTING THREATS USING EVENT PROFILES MAKES


MORE CONVENIENCE TO GET HIGHER ACCURACY IN DETECTING
ADVANTAGES OF PROPOSED SYSTEM

• OUR PROPOSED SYSTEM AIMS AT CONVERTING A LARGE AMOUNT OF


SECURITY EVENTS TO INDIVIDUAL EVENT PROFILES FOR PROCESSING VERY
LARGE SCALE DATA.

• USING DEEP LEARNING ALGORITHMS USING ARTIFICIAL NEURAL


NETWORK(WHICH CAN MIMIC WITH COMPUTER NEURONS), CAN
DETECT DIVIDED INDIVIDUAL EVENT PROFILES EASILY.
SOFTWARE REQUIREMENTS
OPERATING SYSTEM : WINDOWS 10
CODING LANGUAGE : PYTHON (VERSION :3.7.0)

HARDWARE REQUIREMENTS
System : MINIMUM i3

Hard Disk : 512 GB

Ram : 4 GB
THANK YOU

You might also like