Understanding Machine Learning

Introduction: This module introduces the background to Machine Learning - Definition of Machine Learning (ML) - Origins of ML - Rule deduction (Expert Systems) vs induction (ML) - Why do we want machines to learn?

Case studies - Regression as a classic example of ML

Data collection and preparation: Collecting the correct data for the training and testing phases is crucial - The data is often ‘dirty’ and needs to be cleaned - But more than that, the way in which the data is pre-processed is often the difference between poor and highly effective ML - Data for supervised and unsupervised learning - Data sampling - Data volume reduction - Removing ambiguities - Normalisation - Discretisation - Cleansing - Missing values - Outliers - Data and dimensional reduction - Data understanding - Generalisation of hierarchies

Creating or choosing the algorithm: Building a new algorithm for the data modelling (or, as is often done, choosing an existing one) is a vital part of the process - Examples of creating algorithms - The use of data mining algorithms - Classes and examples of data mining

Classic ML algorithms in Plain English: A number of algorithms are used very frequently - This module will look at some examples and explain what they are trying to achieve and how they work - Regression Clustering - Decision trees - Support Vector Machines (SVMs)- Classification - Segmentation - Association - Sequence analysis - Neural nets - Deep Learning

training the data model: Once the algorithm has been built/selected we train it - Selecting the training data - Ratio of training to test data - How to make an unbiased selection - How to use training data to create the model

Testing and improving the data model: Testing is a vital (and complex) part of the process - Confusion matrices - False positives vs False negatives - Measuring efficiency - ROC curves - Overfitting and bias

Introduction to ML in R: R is a well-established open source language with many built-in ML algorithms - This module introduces the language and provides some practical ML work - Introduction to R - Lab ML with R

ML underpinnings: It is much easier to create effective ML systems if we understand (and have a knowledge of) the hierarchically arranged systems that underpin ML - And we also need to understand that AI (Artificial Intelligence) is based on ML, ML on Data mining and so on - AI - ML - Data mining - Statistics - Maths

Combining data models: Any one ML system that we build will have a certain level of efficiency - But we can build a number of different data models and combine them in various ways so that the efficiency of the whole is greater than the sum of the parts - Ensemble - Boosting - Gradient boosting