Market Basket Analysis & Linear Discriminant Analysis with R

This course has two parts. In part 1 Association rules (Market Basket Analysis) is explained. In Part 2, Linear Discriminant Analysis (LDA) is explained. L

--------------------------------------------------

Details of Part 1 - Association Rules / Market Basket Analysis (MBA)

----------------------------------------------------

• What is Market Basket Analysis (MBA) or Association rules
• Usage of Association Rules - How it can be applied in a variety of situations 
• How does an association rule look like?
• Strength of an association rule - 
  1. Support measure
  2. Confidence measure 
  3. Lift measure
• Basic Algorithm to derive rules
• Demo of Basic Algorithm to derive rules - discussion on breadth first algorithm and depth first algorithm
• Demo Using R - two examples
• Assignment to fortify concepts

--------------------------------------------------

Details of Part 2 - Linear  (Market Basket Analysis)

----------------------------------------------------

• Need of a classification model
• Purpose of Linear Discriminant
• A use case for classification
• Formal definition of LDA
• Analytics techniques applicability
• Two usage of LDA 
  1. LDA for Variable Selection
  2. Demo of using LDA for Variable Selection
  3. Second usage of LDA - LDA for classification

• Details on second practical usage of LDA
  1. Understand which are three important component to understand LDA properly
  2. First complexity of LDA - measure distance :Euclidean distance 
  3. First complexity of LDA - measure distance enhanced  :Mahalanobis distance
  4. Second complexity of LDA - Linear Discriminant function
  5. Third complexity of LDA - posterior probability / Bays theorem

• Demo of LDA using R
  1. Along with jack knife approach
  2. Deep dive into LDA outputn
  3. Visualization of LDA operations
  4. Understand the LDA chart statistics

• LDA vs PCA side by side
• Demo of LDA for more than two classes: understand
  1. Data visualization
  2. Model development
  3. Model validation on train data set and test data sets
• Industry usage of classification algorithm
• Handling Special Cases in LDA