AI and ML for Fraud Detection [HITB+ CYBERWEEK 2021]

$3,299.00

Duration

3 days

Delivery Method

hybrid

Level

advanced

Seats Available

20

Duration

3 days

Delivery Method

hybrid

Level

advanced

 

ATTEND IN-PERSON: Onsite in Abu Dhabi

ATTEND ONLINE: Virtual via Zoom and Discord

DATE: 21-23 November 2021

TIME: 09:00 to 17:00 GST/GMT+4

Date Day Time Duration
21 Nov Sunday 0900-17:00 GST/GMT+4 8 Hours
22 Nov Monday 0900-17:00 GST/GMT+4 8 Hours
23 Nov Tuesday 0900-17:00 GST/GMT+4 8 Hours

 


The Real-life Experience – On the final day, trainees get an elusive session to play

with a fraud analytical engine used in current real-world application!



It is estimated that a typical organization loses about 5 percent of its revenues due to fraud each year. In this course, you will learn how machine learning can be used to fight frauds: you will understand when and how to apply supervised learning algorithms to detect fraudulent behavior similar to past ones, as well as unsupervised learning methods to discover novel types of fraudulent activities.

The course will not focus on the mathematics or theory, but on the practical applications: the course will provide a mix of technical and theoretical insights and shows you how to practically implement fraud detection models.

Moreover, during the course you will understand how to deal with the typical challenges of the fraud analytics task (e.g., data scarcity and imbalancing) and will get advice from real-life experience to help you prevent making common mistakes in the fraud detection domain.

Agenda

  • Introduction to fraud detection and Data Exploration and Preprocessing

    - Frauds: Definition and Types - Fraud Detection and Prevention - Big data and Analytics - Fraud analytics process model - Develop a data-driven fraud-detection system - Data Preprocessing Step

  • Fraud detection using unlabeled data

    - Graphical and Statistical Outlier Detection - Clustering - Evaluating and Interpreting Clustering Solutions

  • Fraud detection using labeled data

    - Linear Regression - Logistic Regression - Decision Trees - Neural Networks - Support Vector Machines - Ensemble Methods: Random Forests

  • Evaluating a Fraud Detection Model

    - Splitting Up the Data Set - Performance Metrics

  • The Real-life Experience

    - Demo and hands on a real fraud analytical engine

Why You Should Take This Course

TBA

Who Should Attend

This course is intended for students who have basic experience in the application of data science and machine learning to security-related tasks. Furthermore, the course is also interesting for anyone who wants to understand how to effectively manage and detect frauds by exploiting data science and machine learning techniques.

Key Learning Objectives

  • Get to know the fraud phenomenon: fraud detection, prevention, and analytics.

  • Understanding the basic ingredient of any fraud analytical model, Data: data sources, visual exploration, descriptive statistics, data preprocessing activities and feature engineering techniques.

  • Building, applying, and evaluating machine learning algorithms to identify potential frauds: Unsupervised and supervised learning techniques.
  • Prerequisite Knowledge

    • A basic understanding of data analytics: descriptive statistics (e.g., mean, standard deviation, correlation, conīŦdence intervals, hypothesis testing), data handling (using for example, Microsoft Excel, SQL, etc.), and data visualization (e.g., bar plots, pie charts, histograms, scatter plots, etc.).
    • A basic understanding of machine learning techniques
    • Familiarity with Python

    Hardware / Software Requirements

    • Virtualbox installed, 6/8GB of RAM and 10GB of storage.
    • Students will be provided with a Linux VM containing all necessary tools and setups

    Your Instructor

    Michele received his Ph.D. degree cum laude in Information Technology from Politecnico di Milano in Italy, where he is currently a Contract Professor and a Postdoctoral Researcher working as part of the System Security group inside the Dipartimento di Elettronica, Informazione e Bioingegneria.


    His research revolves around the application of machine learning methods to various cybersecurity-related fields, ranging from cyber-physical and automotive systems to binary analysis, going through fraud and intrusion detection. In particular, his research focuses mainly on the financial fraud detection task, where he worked on the analysis of advanced financial threats such as banking Trojans, on the design of frameworks to identify and timely detect fraudulent transactions, and on the security evaluation of detection systems against adversarial machine learning attacks.

    He is actively involved in research projects funded by the European Union, and he is also co-founder of Banksealer, a Fintech spinoff of Politecnico di Milano.