Akshu MCA Seminar ppt template (1)

1.

PUNE DISTRICT EDUCATION ASSOCIATION’S
College of Engineering
Manjari (BK'), Pune.
Department Of Computer Engineering
Seminar Topic Name : Phishing Attacks and Prevention Techniques
Name : Akanksha Sachin Vyawahare
Roll No. : 1185
Seminar Guide
Prof A. A. Godage

2.

TABLE OF CONTENTS :Problem Statement
Literature Survey
Introduction
Objectives
Existing System
System Architecture
Technologies Used
Working Methodology
Future Scope
Conclusion
References

3.

Problem Statement
• Phishing attacks are increasing rapidly on the internet.
• Fake websites and malicious links are used to steal sensitive user
information.
• Users often cannot differentiate between genuine and phishing websites.
• Traditional detection methods fail to detect newly created phishing websites.
• Phishing attacks can lead to financial loss, identity theft, and data breaches.
• There is a need for an intelligent and automated system to detect phishing
websites in real time.
• The system should provide accurate predictions and warning alerts to
protect users from cyber threats.

4.

Literature Survey
Saleh et al.
Improved Decision Tree,
Random Forest,
(2024)
Random Forest &
Decision Tree,
XGBoost for Telecom
XGBoost
Random Forest improves prediction
accuracy with hyperparameter
tuning
Churn
Alotaibi et al.
Customer Churn Prediction in
Telecom Sector
Random Forest,
Poudel et al.
Explainable ML for
Random Forest +
(2024)
Telecom Churn Prediction
Explainable AI
Aggarwal et al.
(2025)
Telecom Churn Prediction using
ML Models
Random Forest,
RF provides strong baseline
XGBoost, TabNet
performance among ML
models
Abdelhady et
al. (2025)
AI-based Churn Prediction in
Telecom
Random Forest
Achieved ~95% accuracy and high
AUC using RF
Asif et al.
Explainable AI Approach for
Churn Prediction
RF + XAI
(2024)
(2025)
LGBM, ANN
Classifier
Techniques
Ensemble models (RF + boosting)
provide better performance
Improves interpretability and
decision-making
Enhances transparency and feature
importance analysis

5.

Introduction
• The internet is widely used for banking, shopping, education, and
communication.
• Phishing is a cyber attack that uses fake websites, emails, or messages to steal
personal information.
• Traditional detection methods cannot effectively identify new phishing
websites.
• Machine Learning and Artificial Intelligence help detect phishing attacks
automatically.
• These techniques analyze URLs and website features to identify suspicious
websites in real time.
• Phishing detection systems help prevent online fraud, identity theft, and data
breaches.

6.

Objectives
•To understand the concept of phishing attacks and cybersecurity threats.
•To study different types of phishing attacks.
•To analyze how phishing websites trick users.
•To learn phishing detection and prevention techniques.
•To explore the use of Machine Learning in phishing detection.
•To increase cybersecurity awareness among users.
•To reduce online fraud, identity theft, and data breaches.
•To promote safe and secure internet usage.

7.

Existing System
•Uses blacklist-based phishing detection methods.
•Detects only previously known phishing websites.
•Relies on predefined rules and signatures.
•Provides browser security warnings to users.
•Uses manual verification for suspicious websites.
•Requires regular updates of phishing databases.
•Cannot effectively detect zero-day phishing attacks.
•Limited accuracy against advanced phishing techniques.

8.

System Architecture
The Phishing Website Detection System is designed to identify phishing
websites using Machine Learning techniques.
The system analyzes website URLs and predicts whether a website is phishing or
legitimate.
It consists of modules such as URL Input, Data Preprocessing, Feature
Extraction, Machine Learning Detection, Prediction & Alert, and Database
Storage.
System Flow
User enters a website URL → Data Preprocessing → Feature Extraction → Machine
Learning Model Analysis → Phishing Detection → Risk Score & Threat Level
Generation → Result Display → Data Stored in Database.

9.

Technologies Used
•Frontend: HTML, CSS, JavaScript
•Backend: Node.js, Express.js
•Machine Learning: Python, Scikit-learn
•Database: MySQL / MongoDB
•Tools: VS Code, GitHub
•Technology: Machine Learning, Cybersecurity, REST API

10.

Working Methodology
• User enters a website URL into the system.
• Data Preprocessing is performed to clean and validate the URL.
• Feature Extraction extracts important URL features such as:
- URL Length
- HTTPS Status
- Special Characters
- Domain Information
• Machine Learning Model analyzes the extracted features.
• The model predicts whether the website is Phishing or Legitimate.
• The system generates a Risk Score and Threat Level.
• The prediction result is displayed to the user.
• The result is stored in the database for future reference and analysis.

11.

Future Scope
•Improve detection accuracy using advanced Machine Learning and Deep Learning model
•Develop a browser extension for automatic phishing detection.
•Create a mobile application for real-time website scanning.
•Integrate real-time threat intelligence and blacklist databases.
•Add email phishing detection capabilities.
•Implement cloud-based deployment for wider accessibility.
•Enhance risk analysis and threat prediction features.
•Support detection of new and emerging phishing attacks.
•Improve user awareness through security recommendations and alerts.
•Extend the system for enterprise-level cybersecurity solutions.

12.

Conclusion
Phishing attacks are one of the most serious cybersecurity threats in today's
digital world. They can lead to financial loss, identity theft, and data breaches.
Traditional detection methods are often unable to detect newly created phishing
websites effectively.
Machine Learning-based phishing detection provides an intelligent and
automated solution by analyzing website URLs and identifying suspicious
patterns in real time. It improves detection accuracy, enhances user security, and
helps prevent online fraud.
Overall, phishing detection and prevention techniques play an important role in
creating a safer online environment and protecting users from cyber threats.

13.

References
[1].Rehman et al.,
“Real-Time Phishing URL Detection Using Machine Learning,”
International Journal of Cybersecurity and Digital Protection, 2025.
[2]. Albishri et al.,
“Comparative Analysis of Machine Learning Techniques for URL Phishing
Detection,” Journal of Information Security and Applications, 2024.
[3]. Jadhav et al.,
“Hybrid Machine Learning Framework for Phishing Website Detection,”
IEEE Conference on Artificial Intelligence and Cybersecurity, 2025.
[4]. Kibriya et al.,
“Deep Learning Based Phishing Website Detection System,”
International Journal of Advanced Computer Science and Applications,
2025.