Learning Objectives:

  • Define artificial intelligence and its sub-fields: machine learning, deep learning, and natural language processing.

  • Understand the key machine learning algorithms used in banking.

  • Identify concrete AI applications in banking: credit scoring, customer service, and fraud detection.

3.1 What is Artificial Intelligence?

Artificial intelligence is the simulation of human intelligence in machines that are programmed to think, learn, and make decisions. The NUS course provides “Introduction to data science, machine learning, and artificial intelligence concepts” . The Knowledge Academy course covers “fundamentals of Artificial Intelligence and Machine Learning” and “data handling techniques” . The STEP course defines AI as mimicking “the perceiving, cognising and decision-making capabilities of a human being via computing machines intelligence” and notes that “AI is so ubiquitous that you actually use it dozens of times a day without noticing it” .

Key Sub-Fields of AI:

  • Machine Learning: Algorithms that learn from data without explicit programming. The NUS course provides “Detailed discussion on AI algorithms and specific use cases” and “Introduction to Supervised and Unsupervised Learning techniques” .

  • Deep Learning: Neural networks with multiple layers capable of learning complex patterns. The NUS course provides “Overview of Deep Learning and Natural Language Processing” .

  • Natural Language Processing: AI for understanding and generating human language. The STEP course notes applications include “speech recognition” .

3.2 Key Machine Learning Algorithms

The NUS course provides “Detailed discussion on AI algorithms and specific use cases” . The NobleProg course covers “Machine Learning and AI Fundamentals” including “Supervised vs. unsupervised learning” . The Knowledge Academy course covers “predictive analytics, fraud detection, credit scoring, algorithmic trading” .

Key Algorithms:

  • Supervised Learning: Algorithms that learn from labelled data. Applications include credit scoring, fraud detection, and customer churn prediction.

  • Unsupervised Learning: Algorithms that find patterns in unlabelled data. Applications include customer segmentation and anomaly detection.

  • Reinforcement Learning: Algorithms that learn through trial and error. The NUS course includes an “Overview of Reinforcement Learning and its applications” .

Specific Algorithms:

  • Logistic Regression: Used for binary classification (e.g., default vs. non-default).

  • Decision Trees and Random Forests: Used for credit scoring and fraud detection.

  • Gradient Boosting: Used for predictive modeling in credit scoring.

  • Neural Networks: Used for deep learning applications.

  • Clustering Algorithms: Used for customer segmentation.

3.3 AI Applications in Banking

The NUS course provides “Exploration of AI use cases in financial services” . The Amrita course covers “Applications of AI in Banking – Need of AI in Banking – Importance of AI in banking – AI in customer service – AI in risk Management – AI in Fraud detection – Future of AI in Banking” . The STEP course covers “Concrete AI applications in Finance: Digital Identity Verification” and “Concrete AI applications in Finance: Credit Scoring” . The Knowledge Academy course identifies “predictive analytics, fraud detection, credit scoring, algorithmic trading, and automated customer service” as key applications .

Customer Service:

  • The STEP course covers “Digital Identity Verification” as an AI application .

  • Chatbots and virtual assistants provide 24/7 support. The Knowledge Academy course notes AI improves “customer service through chatbots” .

  • Personalisation: The Amrita course covers “Data-driven personalization of banking and insurance products” .

Risk Management and Credit Scoring:

  • The STEP course covers “Concrete AI applications in Finance: Credit Scoring” .

  • The NobleProg course covers “Risk assessment, fraud detection, and anomaly detection” .

  • AI-powered credit scoring models assess creditworthiness using alternative data.

Fraud Detection:

  • The Amrita course covers “AI in fraud detection” .

  • The NobleProg course covers “Fraud detection, and anomaly detection” .

Other Applications:

  • Algorithmic Trading: The Knowledge Academy course identifies “algorithmic trading” as a key application .

  • Regulatory Compliance: The NobleProg course covers “Using analytics to identify trends, patterns, and risks” and “Developing dashboards and reporting tools for regulatory assessments” .

PreviousNext