Machine Learning Masterclass

• This course is for programmers who want develop a career in machine learning and discover new inventions
• This course can be enrolled by those who already learned Python and want to implement in machine learning
• This course has foundational education on machine learning and can be taken by those who are complete beginners
• This course is also open for professional programmers who already started their career in machine learning and want to educate themselves more
• Individuals who have passion towards machine learning and want to commit to a future in this field can join this course.

This Machine learning online has no academic prerequisites and is open to students from all academic disciplines. You will, however, need a laptop, desktop, tablet, or smartphone and a reliable internet connection.

Welcome to the course

• Introduction

Setting up R Studio and R crash course

• Installing R and R studio
• Basics of R and R studio
• Packages in R
• Inputting data part 1: Inbuilt datasets of R
• Inputting data part 2: Manual data entry
• Inputting data part 3: Importing from CSV or Text files
• Creating Barplots in R
• Creating Histograms in R

Basics of Statistics

• Types of Data
• Types of Statistics
• Describing the data graphically
• Measures of Centers
• Measures of Dispersion

Introduction to Machine Learning

• Introduction to Machine Learning
• Building a Machine Learning Model

Data Preprocessing for Regression Analysis

• Gathering Business Knowledge
• Data Exploration
• The Data and the Data Dictionary
• Importing the dataset into R
• Univariate Analysis and EDD
• EDD in R
• Outlier Treatment
• Outlier Treatment in R
• Missing Value imputation
• Missing Value imputation in R
• Seasonality in Data
• Bi-variate Analysis and Variable Transformation
• Variable transformation in R
• Non Usable Variables
• Dummy variable creation: Handling qualitative data
• Dummy variable creation in R
• Correlation Matrix and cause-effect relationship
• Correlation Matrix in R

Linear Regression Model

• The problem statement
• Basic equations and Ordinary Least Squared (OLS) method
• Assessing Accuracy of predicted coefficients
• Assessing Model Accuracy – RSE and R squared
• Simple Linear Regression in R
• Multiple Linear Regression
• The F – statistic
• Interpreting result for categorical Variable
• Multiple Linear Regression in R
• Test-Train split
• Bias Variance trade-off
• Test-Train Split in R

Regression models other than OLS

• Linear models other than OLS
• Subset Selection techniques
• Subset selection in R
• Shrinkage methods – Ridge Regression and The Lasso
• Ridge regression and Lasso in R

Classification Models: Data Preparation

• The Data and the Data Dictionary
• Importing the dataset into R
• EDD in R
• Outlier Treatment in R
• Missing Value imputation in R
• Variable transformation in R
• Dummy variable creation in R

The Three classification models

• Three Classifiers and the problem statement
• Why can’t we use Linear Regression?

Logistic Regression

• Logistic Regression
• Training a Simple Logistic model in R
• Results of Simple Logistic Regression
• Logistic with multiple predictors
• Training multiple predictor Logistic model in R
• Confusion Matrix
• Evaluating Model performance
• Predicting probabilities, assigning classes and making Confusion Matrix in R

Linear Discriminant Analysis

• Linear Discriminant Analysis
• Linear Discriminant Analysis in R

K-Nearest Neighbors

• Test-Train Split
• Test-Train Split in R
• K-Nearest Neighbors classifier
• K-Nearest Neighbors in R

Comparing results from 3 models

• Understanding the results of classification models
• Summary of the three models

Simple Decision Trees

• Basics of Decision Trees
• Understanding a Regression Tree
• The stopping criteria for controlling tree growth
• The Data set for this part
• Importing the Data set into R
• Splitting Data into Test and Train Set in R
• Building a Regression Tree in R
• Pruning a tree
• Pruning a Tree in R

Simple Classification Tree

• Classification Trees
• The Data set for Classification problem
• Building a classification Tree in R
• Advantages and Disadvantages of Decision Trees

Ensemble technique 1 – Bagging

• Bagging
• Bagging in R

Ensemble technique 2 – Random Forest

• Random Forest technique
• Random Forest in R

Ensemble technique 3 – GBM, AdaBoost and XGBoost

• Boosting techniques
• Gradient Boosting in R
• AdaBoosting in R
• XGBoosting in R

Maximum Margin Classifier

• Content flow
• The Concept of a Hyperplane
• Maximum Margin Classifier
• Limitations of Maximum Margin Classifier

Support Vector Classifier

• Support Vector classifiers
• Limitations of Support Vector Classifiers

Support Vector Machines

• Kernel Based Support Vector Machines

Creating Support Vector Machine Model in R

• The Data set for the Classification problem
• Importing Data into R
• Test-Train Split
• Classification SVM model using Linear Kernel
• Hyperparameter Tuning for Linear Kernel
• Polynomial Kernel with Hyperparameter Tuning
• Radial Kernel with Hyperparameter Tuning
• The Data set for the Regression problem
• SVM based Regression Model in R