1. Unsupervised Learning
1.1. Associated analysis
1.1.1. APRIORI
1.1.2. Eclat
1.1.3. FP Growth
1.2. Data without Labels
1.3. Clustering
1.3.1. K- means
1.3.2. K median
1.3.3. Hierarchy clustering
1.3.4. Exception maximization
1.3.5. Example
1.3.5.1. Identifying fake news
1.3.5.2. Document analysis
1.4. Dimensionality reduction
1.4.1. Feature extraction
1.4.1.1. Principle component analysis
1.4.2. Feature selection
1.4.2.1. Wrapper
1.4.2.2. Filter
1.4.2.3. Embedded method
1.4.2.4. Example
2. Performance Evaluation
3. Supervised Learning
3.1. Data with label
3.2. Classification
3.2.1. Logistic regression
3.2.2. Naive Bayes Regression
3.2.3. K nearest neighbor
3.2.4. Support vector machine
3.2.5. Examples
3.2.5.1. Email Spam Detection
3.2.5.2. Speech recognition
3.2.6. Evaluation methods
3.2.6.1. Confusion matrix
3.2.6.1.1. Sample
3.2.6.2. Classification measure
3.2.6.2.1. Accuracy
3.2.6.2.2. Precision
3.2.6.2.3. Recall / Sensitivity
3.2.6.2.4. F1 Score
3.2.6.2.5. False positive rate
3.2.6.2.6. False negative rate
3.3. Regression
3.3.1. Polynomial regression
3.3.2. Linear regression
3.3.2.1. Watch-out
3.3.2.1.1. Outliers
3.3.2.1.2. Correlated features
3.3.2.1.3. Non- linearity
3.3.2.1.4. Collinearity
3.3.2.1.5. Multicollinearity
3.3.3. Ridge regression
3.3.4. Logistic regression
3.3.4.1. Binary
3.3.4.1.1. Application
3.3.4.2. Multinomial
3.3.4.3. Ordinal
3.3.5. Lasso regression
3.3.6. 6. Bayesian Linear Regression
3.3.7. Questions
3.3.7.1. How the regression accuracy measured
3.3.7.2. How is colinearity taken care of?
4. Reinforcement learning
4.1. Model free
4.1.1. Q learning
4.1.2. Hybrid
4.1.3. Policy optimization
4.2. Model based
4.2.1. Learn the model
4.2.2. Given the model
5. Model Maintenance
5.1. Deterioration Factors
5.1.1. Additional features
5.1.2. Behavioral changes
5.1.3. Process changes
5.1.4. Changes in current factors
5.1.5. Competition
5.1.6. Industry changes
5.1.7. Regulation changes
5.1.8. Product / Service changes
5.1.9. Depletion
5.2. Model Maintenance
5.2.1. Assess
5.2.1.1. Check CAP Curve
5.2.1.2. Retrain
5.2.1.3. Rebuild
6. Algorithm Selection
6.1. Predicting Numeric Values
6.1.1. Regression
6.2. Predicting Categories
6.2.1. Classification
6.3. Discover undefined groups
6.3.1. Clustering
6.4. Predict the future
6.4.1. Forecasting
6.5. Understand Key Influencers
6.5.1. Dimesionality reduction