Imperial College Business School Machine Learning

Note: This programme includes a scheduled break for the holiday season. There are no assignments, submissions, or sessions held between 25 December 2020 – 1 January 2021.

Programme highlights

Gain a practical understanding of the tools and techniques used in machine learning applications for business. By the end of this programme, you will be able to:

Characterise the fundamental machine learning problem and outline the ten steps in a typical machine learning project.

Explain why we may not be able to draw meaningful conclusions from experience and calculate the probability of a function providing the correct outcome.

Outline the steps to selecting a machine learning model, select the best fit based on the training set and the validation set and predict a model’s performance.

Differentiate between ranking and prediction problems. Use performance measures to evaluate regression problems, a confusion matrix to evaluate classification problems and lift charts to evaluate ranking problems.

Use oversampling to improve the misclassification rate on interesting cases and the K-fold cross-validation algorithm to overcome shortcomings of the training set-validation set approach.

Understand real-life applications of k-nearest neighbours and use k-nearest neighbours methods for classification and regression.

Apply the Naïve Bayes Theorem to calculate conditional probabilities and explore its real-life applications.

Utilise classification and regression trees to solve real-life problems.

Define proximity for clustering methods and understand the steps involved in hierarchical and k-means clustering and their related applications.

75%

of Netflix users select films recommended to them by the company’s ML algorithms

SOURCE: FORBES, JAN 2020

$20.8B

is the projected global ML market value by 2024

SOURCE: ZION MARKET RESEARCH, NOV 2018

$28.5B

Investment in ML application in Q1 2019

SOURCE: STATISTA, MAY 2019

Download Brochure

Who is this programme for?

This programme is designed for experienced managers and executives working in technology, including:

Mid to senior-level technical managers looking to build a better understanding of machine learning tools and techniques.
Technology management executives seeking to build machine learning capabilities in their function or organisation.
Consultants aiming to develop their knowledge of machine learning to offer better solutions to their clients.

The programme is relevant across industries, including: IT Products & Services, Banking & Financial Services, Healthcare, Consulting, Education, FMCG, Retail, and Telecommunications.

No prior programming experience required.

Modules

Module 1:

Introduction to Machine Learning

What is machine learning, the machine learning process, the machine learning landscape, machine learning in the real world.

Module 2:

The Fundamental Limits of Machine Learning

Is learning feasible at all, interpreting the bound, a probabilistic setting, when is machine learning feasible.

Module 3:

Evaluating Predictive Performance (I)

Which fit is "right", test set, validation set, the "training set - validation set - test set" approach.

Module 4:

Evaluating Predictive Performance (II)

Performance measures for regression, lift charts for classification problems, problems (use confusion matrix), lift charts for regression problems.

Module 5:

Evaluating Predictive Performance (III)

Oversampling, k-fold cross-validation.

Module 6:

K-Nearest Neighbours

K-nearest neighbours for classification, binary and categorical predictors, k-nearest neighbours for regression, distance functions, how should we choose k.

Module 7:

Naïve Bayes

Motivation, exact Bayes classifiers, the Laplace Estimator, Bayes' Theorem, naïve Bayes classifiers.

Module 8:

Classification and Regression Trees

Classification trees, choosing the best split: Part 2, regression trees, random forests and boosting algorithms, choosing the best split: Part 1, pruning a classification tree, bagging.

Module 9:

Cluster Analysis

Motivation, hierarchical clustering is myopic, practical concerns of a cluster, hierarchical clustering, k-means clustering, analysis.

Module 10:

Final Assignment

Module 1:

Introduction to Machine Learning

What is machine learning, the machine learning process, the machine learning landscape, machine learning in the real world.

Module 6:

K-Nearest Neighbours

K-nearest neighbours for classification, binary and categorical predictors, k-nearest neighbours for regression, distance functions, how should we choose k.

Module 2:

The Fundamental Limits of Machine Learning

Is learning feasible at all, interpreting the bound, a probabilistic setting, when is machine learning feasible.

Module 7:

Naïve Bayes

Motivation, exact Bayes classifiers, the Laplace Estimator, Bayes' Theorem, naïve Bayes classifiers.

Module 3:

Evaluating Predictive Performance (I)

Which fit is "right", test set, validation set, the "training set - validation set - test set" approach.

Module 8:

Classification and Regression Trees

Classification trees, choosing the best split: Part 2, regression trees, random forests and boosting algorithms, choosing the best split: Part 1, pruning a classification tree, bagging.

Module 4:

Evaluating Predictive Performance (II)

Performance measures for regression, lift charts for classification problems, problems (use confusion matrix), lift charts for regression problems.

Module 9:

Cluster Analysis

Motivation, hierarchical clustering is myopic, practical concerns of a cluster, hierarchical clustering, k-means clustering, analysis.

Module 5:

Evaluating Predictive Performance (III)

Oversampling, k-fold cross-validation.

Module 10:

Final Assignment

Prerequisite: This programme will require prior knowledge of statistics, probability and linear algebra.
Note: For those wanting to develop deeper skills with analytics, academic credit from this programme can be applied to the Imperial MSc in Business Analytics in the future.

Download Brochure

Faculty

Professor Wolfram Wiesemann

Professor of Analytics and Operations, Imperial College Business School

Wolfram Wiesemann is Professor of Analytics and Operations at Imperial College Business School, London, where he also serves as the Academic Director of the MSc Business Analytics programme as well as a Fellow of the KPMG Centre for Advanced Business Analytics... More info

Certificate

Upon completion of the programme, participants will be awarded a verified Digital Certificate by Imperial College Business School Executive Education.

Future learning:

For those who want to progress their skills to the next level, an academic credit from this programme can be applied to the Imperial MSc in Business Analytics in the future. The MSc programme enables graduates to understand the challenge of managing large data sets and to provide them with a skill set to meet this challenge. The programme combines academic rigour and practical relevance. To learn more, visit the programme website.

Download Brochure

Imperial Machine Learning for Decision Making

Get Your Brochure

STARTS ON

DURATION

PROGRAMME FEE

Special pricing for group enrolments

Welcome

Programme highlights

75%

SOURCE: FORBES, JAN 2020

$20.8B

SOURCE: ZION MARKET RESEARCH, NOV 2018

$28.5B

SOURCE: STATISTA, MAY 2019

Who is this programme for?

Modules

Module 1:

Module 2:

Module 3:

Module 4:

Module 5:

Module 6:

Module 7:

Module 8:

Module 9:

Module 10:

Module 1:

Module 6:

Module 2:

Module 7:

Module 3:

Module 8:

Module 4:

Module 9:

Module 5:

Module 10:

Faculty

Professor Wolfram Wiesemann

Professor of Analytics and Operations, Imperial College Business School

Professor Wolfram Wiesemann

Certificate

Certificate

Early registrations are encouraged. Seats fill up quickly!