Enhance cybersecurity with our cross-platform phishing detection and classification solution. Safeguard your digital environment by identifying and categorizing phishing threats across diverse platforms, ensuring comprehensive protection against malicious activities. Stay ahead in the ever-evolving landscape of cyber threats with our advanced, multi-platform defense mechanism.
IAC 2024 - IA Fast Track to Search Focused AI Solutions
Seminar On.pdf
1. Government College of Engineering, Amravati
2023-24
Cross-platform Phishing Detection and
Classification
Seminar On
Presented By -Tejas Lakade
Information Technology Department
Seminar Guide
D.R.Uike
Head of Department
Prof. A. W. Bhade
2. DETECT AND CLASSIFY PHISHING ATTACKS
Develop methods and techniques to accurately
identify and classify phishing attacks across multiple
platforms.
ENHANCE CYBERSECURITY
Provide effective countermeasures to improve the
overall cybersecurity posture and protect users from
malicious phishing attempts.
ENABLE PROACTIVE RESPONSE
Enable organizations to proactively respond to
evolving phishing techniques by staying ahead of
attackers.
Aim and
Objectives
3. Background and Motivation
RISING THREAT OF PHISHING
Phishing attacks caused
$21.6 million loss between
January & June 2012
For Q1 2022, LinkedIn was
the most imitated brand for
phishing attempts globally
45.56% of emails sent in
2021 were spam
USER VULNERABILITY
Phishing sites mimic real
ones, making it tough to spot
the difference in designs
False urgency in phishing
leads users to rush actions,
overlooking authenticity and
risking errors
NEED FOR CROSS-PLATFORM
SOLUTION
Provide consistent defense
across smartphones, tablets,
computers, and IoT for
comprehensive protection.
Maintaining proactive
security against emerging
threats on multiple platforms.
4. What is Phishing ?
A phishing attack is a method of tricking
users into unknowingly providing
personal and financial information or
sending funds to attackers.
The most common form is to use email
to provide a link to what appears to be a
legitimate site but is actually a malicious
site controlled by the attacker.
7. Solution And Approaches
Classification-Based Algorithms and Various Solutions
Various solutions
Decision tree and random forest
Naive Bayes
Neural Networks
Support vector machine (SVM)
Why SVM is suitable
Effective in high dimensional spaces
Robust over
Optimize margin between classes
Works well with limited data
10. EMAIL PLATFORM
Spam Filtering
Enterprise Email Security
Applications
WEB BROWSER
Browser Extension
Security Suites
SOCIAL MEDIA
Link Scanning
Profile Safety Measure
MOBILE APPLICATION
Mobile Security Apps
In-build Browser Security
11. Future Scope
In the future, cross-platform phishing
detection will advance through
ensemble learning, combining varied
features from URLs, content, and
network behaviors. Deep learning
models, like convolutional and recurrent
neural networks, will ensure real-time
threat identification across devices.
Behavioral analysis and adaptable
systems will swiftly evolve against
emerging phishing strategies, supported
by user-friendly interfaces offering
instant alerts.
12. Conclusion
Phishing will never be completely eradicated.
However, a combination of good organization
and practice, proper application of current
technologies, and improvements in security
technology has the potential to drastically
reduce the prevalence of phishing and the losses
suffered from it.User education remains the
strongest and at the same time, the weakest link
to phishing counter measures.
13. References
Abdelhamid, N., Ayesh, A., & Thabtah, F. (2014). Phishing detection based Associative Classification data
mining. Expert Systems with Applications, 5948-5959.
Akanbi, O. A., Amiri, I. S., & Fezaldehkordi, E. (2015). A Machine Learning Approach to Phishing Detection
and Defense. ELSEVIER.
Anti-Phishing Working Group, J. (2017, Feb. 23). Phishing Activity Trends Report, 4th Quarter 2016.
Retrieved March 10, 2017, from APWG: https://docs.apwg.org/reports/apwg_trends_report_q4_2016.pdf
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning. 20(3): 273-297.
Fang, X., Koceja, N., Zhan, J., Dozier, G., & Dipankar, D. (2012). An Artificial Immune System for Phishing
Detection. IEEE World Congress on Computational Intelligence.
Jain, A. K., & Gupta, B. B. (2016). Comparative Analysis of Features Based Machine Learning Approaches
for Phishing Detection.
International Conference on Computing for Sustainable Global Development (INDIACom), (pp. 2125-
2130).
14. Do you have any
questions?
We hope you learned something new.