Scribd
Upload
2 views

Lecture01

The document outlines a lecture on Machine Learning by Arslan Ali Khan, detailing his qualifications and the course contents, which cover topics such as AI, cyber security, machine learning basics, supervised and unsupervised learning algorithms, anomaly detection, malware analysis, and network traffic analysis. It also includes a breakdown of marks distribution for the course and highlights the applications of AI in various fields, particularly in security. The lecture emphasizes the importance of machine learning in making intelligent decisions and predictions based on data.

Uploaded by

usamasulemanleghari
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
2 views

Lecture01

The document outlines a lecture on Machine Learning by Arslan Ali Khan, detailing his qualifications and the course contents, which cover topics such as AI, cyber security, machine learning basics, supervised and unsupervised learning algorithms, anomaly detection, malware analysis, and network traffic analysis. It also includes a breakdown of marks distribution for the course and highlights the applications of AI in various fields, particularly in security. The lecture emphasizes the importance of machine learning in making intelligent decisions and predictions based on data.

Uploaded by

usamasulemanleghari
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

Lecture

Machine Learning 01

Introduction

Arslan Ali Khan


[email protected]
Department of Cyber-Security and Data Science
Riphah Institute of Systems Engineering (RISE),
Riphah International University, Islamabad, Pakistan.
Arslan Ali Khan
Qualifications:
BS(Telecommunication Engineering)
FAST Islamabad

MS(Information Security)
Riphah Institute of Systems Engineering, Islamabad

PhD(Cyber Security) – In Progress


Air University, Islamabad

CHFI
EC-Council

PCAP - Certified Associate in Python Programming


Python Institute

Experience:
Co-Founder
Outclick Labs

Founder
SmarTech Solutions

GARC Consultant - Lecturer


Riphah Institute of Systems Engineering, Islamabad
Tentative Course Contents
Why Artificial Intelligence and Cyber Security?
Cyber Threat Landscape, Cyber Economy, real world application of machine learning and deep learning, limitations of AI
Machine Learning Basics
Training, testing and validation- cleaning, handling imbalanced and noisy data, data splitting, normalization. Overfitting & Under Fitting.
Feature Selection. Classification and parameter tuning of supervised classification algorithms.
Using Supervised Learning Algorithms
Logistic Regression, Decision Tree, Support Vector Machine, Neural Networks, Naïve Bayesian, K-Nearest Neighbour algorithms.
Using Unsupervised Learning Algorithms
K-Means clustering, Hierarchical clustering, DBSCAN algorithms. Choosing Thresholds and Comparing Models, Ensemble Learning.
Using Deep Learning algorithms
Using Multi-layer perceptron (MLP), training deep neural networks, Convolution Neural Network (CNN), Recurrent Neural Networks (RNN),
Processing sequences using CNN and RNN.
Anomaly Detection
Feature engineering for anomaly detection, algorithms, risks and mitigation approaches, challenges, real world examples of anomaly
detection-problems and solutions
Malware Analysis
Understanding malware, feature generation, static and dynamic analysis approaches, classification using machine learning. Malware
classification using deep learning, Android malware category and family classification algorithms
Cryptology and AI
Ensuring privacy in training and classification phases of machine learning, developing non-tamperable and bias-free models,
cryptographic tools & model, tracing unauthorized use of data. Developing cryptographic friendly ML algorithms.

Forensic Analysis using AI-I


Automated Log data classification and clustering, Image data representation using AI
Forensic Analysis using AI-II
Audio/Video classification, anomaly detection and criminal profiling using AI
Network Traffic Analysis using AI-I
Theory of network defence (Access control, Intrusion detection, botnets, honey pots), Data capturing, feature engineering, classification.
Building proactive model to detect cyber-attacks.
Network Traffic Analysis using AI-II
3
Phishing attack detection using ML, Spam detection using ML, ransomware attack detection using machine learning, botnet detection,
Using ensemble models for improved performance
Adversarial Machine Learning
Importance, vulnerabilities in ML algorithms. Attack techniques (model poisoning and evasion attack)
Recommended Readings

4
Marks Distribution of Course
• Assignments 15%

• Project 20%

• Midterm 25 %

• Final 40%

5
Artificial Intelligence
Machine Learning

AI ≠ML
or
AI =ML

7
Machine Learning
• Try to make “intelligent” decisions in the way that humans
do, by drawing generalizations and distilling concepts from
complex information sets without explicit instructions.
• Machine learning refers
 Algorithms and processes that “learn” in the sense of being able to
generalize past data and experiences in order to predict future
outcomes.
 A set of mathematical techniques, implemented on computer
systems, that enables a process of information mining, pattern
discovery, and drawing inferences from data.
Machine Learning
• Designing algorithms that ingest data and learn a
(hypothesized) model of the data

• The learned model can be used to:


 Detect patterns/structures/themes/trends etc. in the
data
 Make predictions about future data and make decisions.
Machine Learning
• Supervised machine learning methods adopt a Bayesian
approach to knowledge discovery, using probabilities of
previously observed events to infer the probabilities of new
events on labelled dataset.
• Unsupervised machine learning methods draw
abstractions from unlabeled datasets and apply these to
new data.
• Both families of methods can be applied to problems of
classification (assigning observations to categories) or
regression (predicting numerical properties of an
observation)
Machine Learning
Problem
Machine Learning
Problem
AI in Real World
Machine Learning
Applications
• Speech Recognition
• Virtual Personal Assistants
• Videos Surveillance
• Fraud Detection
• Playing Games
• Online Customer Support
• Spam Detection
• Natural Language Processing
• Product Recommendations
AI Applications-
Computer Vision
AI Applications-
NLP
AI Applications
AI Applications
In Security
• Malware Detection and Classification
• DDOS attack detection
• Cryptography and AI
• Network Traffic Analysis
• Insider Threat Detection
• Anomaly Detection
• Botnet and Spam Detection
• Digital Forensics
• Phishing Attack detection
Thank You!

You might also like