You are on page 1of 9

CYBER SECURITY THREATS

DETECTION USING ML
CONTENTS:

 Introduction
 Objective
 Literature Survey
 System Requirements Specification
 Summary
INTRODUCTION

 In today's digital era, where individuals and organizations heavily rely on online platforms for
communication and transactions, the menace of phishing attacks has emerged as a pervasive threat to
cybersecurity.

 Phishing, a deceptive practice employed by cybercriminals to trick users into divulging sensitive
information, exploits the trust users place in seemingly legitimate websites.

 As phishing techniques evolve in sophistication, traditional security measures often fall short in providing
adequate protection.
OBJECTIVE

 Phishing, a malicious practice of deceiving individuals into revealing sensitive information, poses
a significant threat to cybersecurity.

 With the increasing sophistication of phishing attacks, there is a growing need for robust and
efficient detection mechanisms.

 This research focuses on developing a phishing detection system using machine learning
classification techniques.
LITERATURE SURVEY

EXISTING SYSTEM:
The existing phishing detection mechanisms often overlook the dynamic temporal patterns
of phishing attacks and do not adequately consider user-centric aspects, leading to a gap in
comprehensive phishing threat identification.

DISADVANTAGES :
1. Lack of comprehensive analysis of temporal patterns in phishing attacks.
2. Insufficient consideration of user-centric factors, such as user behavior and decision-making
processes, in existing phishing detection models.
3. Limited research on the impact of visual similarity analysis in detecting sophisticated phishing
websites.
PROPOSED SYSTEM:
The proposed phishing detection system contributes to bolstering cybersecurity defenses,
providing a proactive approach to identify and mitigate phishing threats, thereby safeguarding
individuals and organizations from falling victim to malicious online activities. Application of ML
techniques, specifically neural networks(CNN), for feature extraction and classification in phishing
website detection.

ADVANTAGES:
1. To analyze temporal patterns of phishing attacks over an extended period.
2. To incorporate user-centric factors, such as behavioral analysis, into the phishing detection
model.
3. To evaluate the effectiveness of visual similarity analysis in identifying sophisticated phishing
websites
SYSTEM REQUIREMENTS SPECIFICATION

HARDWARE REQUIREMENTS:
1. Server with a minimum of 8 GB RAM.
2. Storage capacity of at least 500 GB.

SOFTWARE REQUIREMENTS:
1. Operating System: Linux-based distribution
2. Programming Language: Python 3.x
3. Machine Learning Frameworks: TensorFlow and scikit-learn
4. Database: MongoDB
5. Web Framework: Django
SUMMARY

 This research endeavors to address the pressing need for effective phishing detection by
leveraging the power of machine learning classification. By harnessing the capabilities of
advanced algorithms and techniques, we aim to develop a robust system capable of
discerning between legitimate and phishing websites.

 Phishing, a malicious practice of deceiving individuals into revealing sensitive information,


poses a significant threat to cybersecurity. With the increasing sophistication of phishing
attacks, there is a growing need for robust and efficient detection mechanisms. This research
focuses on developing a phishing detection system using machine learning classification
techniques.
THANK YOU

You might also like