AI for Suspicious Activity Monitoring

Build AI-Powered Systems to Detect Anomalies, Fraud, and Unusual Patterns in Real-Time Using Machine Learning & Gen AI

What you'll learn

• Learn about the uses of self-supervised machine learning
• Implement self-supervised machine learning frameworks such as autoencoders using Python
• Learn about deep learning frameworks such as Keras and H2O
• Learn about Gen AI and LLM Frameworks

Requirements

• Basic Python data science concepts
• Basic Python syntax
• Understanding of the Colab environment
• Introduction to the Gen AI Ecosystem

Description

Unlock the power of AI to detect anomalies, fraud, and suspicious behaviour in digital systems. "AI for Suspicious Activity Monitoring" is a hands-on, end-to-end course designed to teach you how to use traditional AI techniques, deep learning, and generative AI (GenAI) to monitor and respond to unusual patterns in real-world data.

Whether you're a developer, data analyst, or aspiring AI professional, this course provides practical tools and strategies to build intelligent monitoring systems using Python, autoencoders, and large language models (LLMs).

What You’ll Learn

• Anomaly Detection Techniques: Implement classical and modern methods, including statistical outlier detection, clustering-based approaches, and autoencoders.
• Deep Learning for Behaviour Monitoring: Use unsupervised learning (e.g., autoencoders) to detect irregular patterns in time series, text, or sensor data.
• GenAI & LLM Integration: Explore how large language models like OpenAI’s GPT and frameworks such as LangChain and LLAMA-Index can assist in monitoring human-generated activity (e.g., suspicious conversations, document scans).
• Fraud and Cyber Threat Detection: Apply AI tools to detect threats in finance, cybersecurity, e-commerce, and other high-risk domains.
• Cloud-Based Implementation: Build scalable pipelines using tools like Google Colab for real-time or batch monitoring.
• Text Analysis for Audit Trails: Perform NLP-based extraction, entity recognition, and text summarisation to flag risky interactions and records.

Why Enrol in This Course?

In today’s fast-paced digital world, AI-powered monitoring systems are essential to detect threats early, reduce risk, and protect operations. This course offers:

• A practical, Python-based curriculum tailored for real-world applications
• Step-by-step project-based learning guided by an instructor with an MPhil from the University of Oxford and a PhD from the University of Cambridge
• A rare combination of AI, deep learning, and GenAI in a single course
• Use of cutting-edge LLM frameworks like OpenAI, LangChain, and LLAMA-Index to expand beyond numerical anomaly detection into text-based threat detection

Who this course is for:

• Data Scientists who want to increase their knowledge of self-supervised machine learning
• Students of Artificial Intelligence (AI) and Gen AI
• Students interested in learning about frameworks such as autoencoders