Course Description
This course will explain all the following:
What is Feature Engineering?
Features in Machine Learning
Transformation and Extraction of Features
Model Selection: A Quick Review
Linear Sparse Models
Information Theory Meets Machine Learning
Discretization of Numerical Features
Dimensionality Reduction
What am I going to get from this course?
- A comprehensive overview on Feature Engineering strategies including ones you may even have not heard before.
- A practical hands-on style of learning for theoretical concepts.
- A rich and comprehensive introduction to proper references including literature, keywords and notable related scientists to follow.
- Exploring cons & pros and hidden tips on algorithms in practice.
Prerequisites and Target Audience
What will students need to know or do before starting this course?
Background in Math
- Linear Algebra
- Probability Theory & Statistics
- Discrete Mathematics
- Calculus
Background in Machine Learning
- Supervised & Unsupervised Learning
Familiarity with Python Programming
- Sci-kit Learn, Numpy, Matplotlib, etc.
Who should take this course? Who should not?
People interested in a career in Machine Learning can take this course, but should have all the prerequisites.
Curriculum
Module 1: What is Feature Engineering
Lecture 2
You Do Feature Engineering Every Day!
Lecture 3
Features Are Not Always There!
Lecture 4
Underfitting, Fitting, and Overfitting
Lecture 5
Learning in High-Dimensional Setting 1
Lecture 6
Learning In High-Dimensional Setting 2
Lecture 7
Leanring in High-Dimensional Setting 3
Module 2: Data Preparation
Lecture 9
Handling Missing Values - Imputation
Lecture 10
Handling Missing Values - Imputation 2
Lecture 11
Handling Missing Values - Imputation 3
Lecture 12
Feature Transformation - Diversity of Features
Lecture 13
Discretization of Numerical Features
Lecture 14
An Information Theoretic Method
Lecture 15
An Information theoretic Method 2
Lecture 16
Feature Scaling
Module 3: Feature Selection
Lecture 17
Why Feature Selection
Lecture 19
Numerical-Numerical Association
Lecture 20
Spearman Rank Correlation
Lecture 21
Maximal Information Coefficient
Lecture 22
Categorical-Categorical Association
Lecture 23
Categorical Dependence
Lecture 24
Numerical-Categorical Association
Lecture 25
F-Value for Classification
Lecture 26
Sparse Linear Models
Lecture 27
Step-Wise Feature Selection
Lecture 28
A Note on Model Selection
Lecture 29
Ridge Regression
Module 4: Dimension Reduction
Lecture 30
Principal Component Analysis
Lecture 31
Nonnegative Matrix Factorization 1
Lecture 32
Nonnegative Matrix Factorization
Lecture 33
Locally Linear Embedding
Lecture 34
Locally Linear Embedding - Part 2
Module 5: Feature Extraction
Lecture 35
Working With Text Data
Lecture 37
Text Preprocessing
Lecture 38
Text to Numbers 1
Lecture 39
Text to Numbers 2
Lecture 40
Text to Numbers 3
Lecture 41
Working With Time Series
