Predictive Analytics

Supervised Learning Techniques

• Defining supervised learning
• Types of supervised learning:
  
  

•      Regression analysis
•      Classification problems

• Understanding Regression Analysis algorithms:
  
  

•      Linear regression
•      Polynomial regression
•      Logistic regression
  
  

• Understanding Classification problems:

•      Decision tree and random forests algorithms
•      Support vector machine (SUM) algorithm 

• Metrics to evaluate regression and classification models

Unsupervised Learning

• Defining unsupervised learning
• Types of unsupervised learning
  
  

•      Clustering algorithm
•      Dimensionality reduction algorithm
  
  

• Understanding Clustering algorithm:
  
  

•      K- means clustering algorithm
•      Density-based clustering algorithm
•      Hierarchical clustering algorithm
  
  

• Understanding Dimensionality reduction and its application
  
  

•      Principles component analysis (PCA)
•      Singular value decomposition (SVD)
•      t-SNE
  
  

•  Metrics to evaluate unsupervised leaning techniques

Advanced Analytics Techniques

• Identifying ensemble learning methods:
  
  

•     Bagging
•     Random forests
•     Boosting algorithms
•     Stacking & voting classifiers
  
  

• Utilizing XGBoost and LightGBM for gradient boosting
• Defining deep learning concepts and the role of artificial neural networks
• Exploring deep learning framework:
  
  

•     TensorFlow
•     Keras
•     PyTorch
  
  

• Building deep learning:

•     Convolutional neural networks (CNN)
•     Recurrent neural networks (RNN)

Model Selection & Optimization 

• Determining the importance of selecting the best model
• Cross-validation technique to select the best model
• Exploring model bias-variance trade-off
  
  

•        Overfitting
•        Underfitting
  
  

• Course recap and Q&A
• Final project presentation