7.1 -Supervised Learning Continuation

Week 7

Objectives

• Data Preprocessing:
  • Why
  • Common Tasks
    • Cleaning:
      • Missing Data :
        • Ignore
        • Imputation
      • Noisy Data :
        • Binning
        • Regression
        • Clustering
    • Integration
    • Transformation
    • Reduction :
      • Feature Selection :
        • correlation analysis
        • mutual information
        • principal component analysis (PCA)
      • Feature Extraction
      • Sampling
      • Clustering
      • Compression
    • Discretization
    • Normalization
    • Encoding:
      • Ordinal
      • Nominal
        • One hot encoding
        • Label encoding
• Curse of dimensionality
• Machine Learning Models :
  • Supervised :
    • Regression : Revision Linear Regression
      • Regularisation for Feature Selection
      • Best subset, and sequential subsets of features
    • Classification : Naive Bayes
  • Unsupervised :
    • Clustering:
      • Definition
      • Types
      • Kmeans algorithm:
        • Choosing k:
          • Elbow
          • Silhouette
        • Kmeans++

Materials

• Slides
• Jupyter Notebook

Lecture

Next Class