Course Code: mlbanking
Duration: 21 hours
Prerequisites:
  • Experience with Python programming
  • Basic familiarity with statistics and linear algebra is recommended
Overview:

In this instructor-led, live training, participants will learn how to apply machine learning techniques and tools for solving real-world problems in the banking industry.

Participants first learn the key principles, then put their knowledge into practice by building their own machine learning models and using them to complete live team projects.

Course Outline:

Introduction

  • Machine Learning in Finance and Banking

Approaches and Machine Learning Types

  • Statistical learning vs. Machine learning
  • Supervised learning vs unsupervised learning
  • Iteration and evaluation
  • Bias-variance trade-off
  • Languages, libraries and frameworks (Python vs R vs Matlab)

Applied Machine Learning Case Studies

  • Consumer data and big data
  • Consumer and business lending: assessing risk
  • Customer service through sentiment analysis
  • Fraud detection: Beyond conventional security assessment

Hands-on: Python for Machine Learning

  • Preparing the Development Environment
  • Python machine learning libraries and packages
  • Working with scikit-learn and PyBrain

Modeling Business Decisions with Supervised Learning

  • Regression
  • Classification
  • Decision Trees
  • Random Forests
  • Model Evaluation

Regression

  • Linear regression
  • Generalizations and Nonlinearity
  • Exercises

Classification

  • Bayesian refresher
  • Naive Bayes
  • Logistic regression
  • K-Nearest neighbors
  • Exercises

Hands-on: Building an Estimation Model

  • Assessing lending risk based on customer type and history

Cross-validation and Resampling

  • Cross-validation approaches
  • Bootstrap
  • Exercises

Modeling Business Decisions with Unsupervised Learning

  • K-means clustering
  • Examples
  • Challenges of unsupervised learning and beyond K-means
  • Exercises

Hands-on: Building a Recommendation System

  • Building a recommendation system based on customer behavior and activity

Introduction to Neural Networks and Deep Learning

  • Layers and nodes
  • Convolutional Neural Networks
  • Recurrent Neural Networks
  • Multilayer perceptrons
  • Frameworks: Theano, TensorFlow, Keras
  • Developing models in the cloud
  • A word on GPUs

Closing Remarks