Data Mining 2

Feature Engineering

Last semester you learnt some data preparation techniques, such as scaling and transformations. Now we will build on that by covering techniques to generate new features from existing features.

Introduction to Feature Engineering
A short introduction to feature engineering and related tasks.
- Need for feature engineering
- Generic feature engineering steps
- Handout
- Presentation
Feature Selection
The need for feature selection (Curse of Dimensionality) and feature selection techniques
- Curse of Dimensionality
- Feature selection
- Handout
- Presentation
Automatic Feature Engineering
An overview of featuretools, a python module for automatic feature engineering.
- Walk-through of Featuretools
Dimensional Reduction
In this lecture we will cover dimensional reduction techniques. Dimensionality reduction is about representing the data in a lower dimensional space in such a way that certain properties of the data are preserved as much as possible.
- Feature selection vs combination
- Principal Component Analysis (PCA)
- Linear Discriminant Analysis (LDA)
- t-Distributed Stochastic Neighbour Embedding (t-SNE)
Practical 02 - Churn - Baseline Model
In this practical we (you) are going to analyse a churn dataset. The purpose of this practical is to ensure you are happy with both data mining workflow and classification models that were covered in Data Mining 1 module.
- Open
Practical 03 - Churn - Feature Engineering
Short practical on feature engineering
- Open

Feature Engineering

Introduction to Feature Engineering

Feature Selection

Automatic Feature Engineering

Dimensional Reduction

Practical 02 - Churn - Baseline Model

Practical 03 - Churn - Feature Engineering