Scribd
Upload
21 views13 pages

Fainal Project

The presentation by Md. Anik Hosan explores the current impact and future prospects of AI in cybersecurity, emphasizing its role in threat detection and risk assessment. It includes a comprehensive literature review, methodology for comparing AI models, and analysis of results, highlighting the strengths and limitations of various AI applications. The conclusion stresses the importance of addressing challenges to enhance AI-driven cybersecurity innovations.

Uploaded by

md anik hasan
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
21 views13 pages

Fainal Project

The presentation by Md. Anik Hosan explores the current impact and future prospects of AI in cybersecurity, emphasizing its role in threat detection and risk assessment. It includes a comprehensive literature review, methodology for comparing AI models, and analysis of results, highlighting the strengths and limitations of various AI applications. The conclusion stresses the importance of addressing challenges to enhance AI-driven cybersecurity innovations.

Uploaded by

md anik hasan
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/ 13

Welcome to my

presentation

Presented By: Md. Anik Hosan


ID:2025512807
Department of Global IT Engineering
Topics: AI’s Current Impact and Future
Prospects in Cybersecurity
Outline
1.Introduction

2. Literature Review

3. Methodology

4. Result Analysis

5. Challenges and Considerations

6. Conclusion
Introduction

• Importance of Cybersecurity: As ICT systems grow,


protecting data from cyber threats becomes critical.
• Role of AI in Security: AI boosts threat detection, risk
assessment, and response speed.
• Research Focus: This study explores how AI technologies
transform cybersecurity protocols.
• Gaps in Current Research: Limited analysis exists comparing
AI applications' effectiveness.
• Comparative Analysis: The research evaluates AI tools across
security sectors and industries.
Objectives

• Conduct a comprehensive review of AI-based


cybersecurity strategies, focusing on threat detection,
network protection, and prevention of phishing and
fraud.

• Compare various AI models to analyse their


applications, benefits, and limitations in
cybersecurity.
Literature Review
Reference Author Focus Method Findings

111 I.H.Sarker Investigating how AI machine learning (ML), integration AI into


can intelligently and deep learning (DL), natural cyber security
et al. automatically language processing (NLP)
automate
cybersecurity
procedures
48 Hariharan the concept of XAI in machine learning (ML), the importance of
et al. cybersecurity decision-making. explain ability and
accuracy in artificial
intelligence models
77 Malatji and comprehensive adversarial machine highlights AI’s dual
Tolah framework addressing learning (ML) role in both
adversarial and mitigating and
offensive AI strategies facilitating
in cyber security cyberattacks
90 Okdem and real-time threat random forests and high accuracy in
Okdem detection. convolutional neural identifying anomalies
networks (CNNs) and malware
patterns
96 Pritee et al. adaptive recurrent neural networks systems dynamically
authentication (RNNs) evaluate user
Methodology
Key comparisons of AI models in threat detection and
prevention.
Model Use Case Strengths Limitations

Random Threats Handles large Slow for very


Forest classification dataset large dataset .

Convolutional Malware Excellent at Requires large


Neural detention image-like data labeled
Network(CNN analysis. datasets.
s)
Anomaly Identifying Detects novel False positives
Detection zero-day threats without in dynamic
Algorithm threats lebeling. environments
Key comparisons of AI models in network security

Model Use Case Strengths Limitations

Anomaly Identify Effective at High false


Detection unusual finding out- positive in
network liers in nosy
patterns unlabeled environment
data.
Convolution Analyze Detects Computation
al Neural packet data evolving ally
Network(CN threats expensive.
Ns)
Recurrent Monitoring Excels at Required
Neural Net- network pattern preprocessin
works traffic flow recog- nition g for
(RNNs) in structured formats.
inputs
RESULT ANALYSIS

Domain AI Model Accu Prec Re F1- Souce


racy esio call Score
(%) n
Threat Random 99.83 99.86 99. 99.83 PLOS
Detectio Forest 82 ONE
n and Autoenco (2024)
Preventi der [28]
on
Network LSTM- 95.74 95.8 96. 96.0 arXiv
Security Based 2 (2024)
Hybrid [50]
Model
Challenges and Considerations

Domain Limitations Research Directions


Threat Computational Investigation of novel AI
Detectio Complexity techniques, Address
n Data privacy Scalability and
concerns capabilities.
Network Protocol Exploration of real time
Security coverage threat detection capability
limitations - Utilization of diverse
Real-world data sources.
applicability
Conclusion

• AI is making cybersecurity smarter, faster, and stronger.


• Different AI tools are powerful but still face problems like cost
and cyber-attacks.
• Solving current challenges is key to building better security
with AI.
• This study helps set the path for future AI-powered
cybersecurity innovation
Reference

Toufiq, I.I. and Bappi, N.I., 2025. A Data-Driven Approach to Cybersecurity: Advanced Analytics for Identifying
and Preventing Security Breaches. Intelligent Data Science and Analytics, 1(01), pp.39-54.
Camacho , N. G. . (2024) “The Role of AI in Cybersecurity: Addressing Threats in the Digital Age”, Journal of
Artificial Intelligence General science (JAIGS) ISSN:3006-4023, 3(1), pp. 143–154. doi: 10.60087/jaigs.v3i1.75.
de Silva de Alwis, R., 2025. Gendering the New International Convention on Cybercrimes and New Norms on
Artificial Intelligence and Emerging Technologies. Washington Journal of Law, Technologyy & Arts, p.1.
Sebestyen, H., Popescu, D.E. and Zmaranda, R.D., 2025. A Literature Review on Security in the Internet of
Things: Identifying and Analysing Critical Categories. Computers, 14(2), p.61.

You might also like