Share this link via
Or copy link
Introduction and Motivation for Data Science
Basic Statistical Concepts
Central Tendencies
Measures of variability
Probability Basics – Rules and Types
Conditional Probability and Bayes’ Theorem
Probability Distributions
Bernoulli
Geometric
Binomial
Poisson
Exponential
Normal Distribution – Continuity Correction
Sampling Distribution of Means - Central Limit Theorem
Confidence Levels and Intervals
Hypothesis Testing
t-distribution
z-distribution
Chi-squared distribution
F distribution
Annova
Errors
Linear Regression
Simple Linear Regression
Covariance and Correlation
Coefficient of Determination
Residual Analysis
Interpreting Residuals
Testing the model
R-Squared and Significance – Caveats
Leverage
Multiple Linear Regression
Interpreting Regression Coefficients
Assumptions of Multiple Linear Regression
Adjusted R-Squared
Handling Simple Non-linearity
Feature selection & Model building
Multicollinearity
Evaluating the accuracy of forecasts
Logistic Regression & Naïve Bayes Classification
Introduction
Probability vs Likelihood
Maximum Likelihood Estimation
Odds Ratio
Psedo R2
Kappa Metric
ROC Curves and AUC
Gains and Lifts Charts
Naïve Bayes Algorithm
Classification problems with multiple classes
Naïve Bayes Assumptions
PCA & Regularization
Dimensionality Reduction
Principal Component Analysis
Methodology
Evaluating Model Accuracy
Error Measures
Bias-Variance Trade-off
Overfitting vs Underfitting
Regularization
Ridge Regression
Lasso Regression
Elastic-net
Time-series Forecasting
Time-series data
Forecasting
Regression on Time
Seasonality
Auto-regressive methods
Components of a time-series
Autocorelation (ACF )and Partial-Autocorelation (PACF)
Stationary and non-stationary data
SMA, WMA & Exponential smoothing
Holt-Winters Model
Residual Plots
AR, MA and ARIMA models
Model Selection
Unsupervised Models – Clustering
Understanding Distance
Distance between Categorical Attributes
Value Distance Measure
Kernel Tricks
Clustering Applications
Hierarchial (Agglomerative) clustering
K-means and K-medoids
Stopping or Convergence Criteria
Stability Check of Clusters
Association Rules
Frequent Pattern Analysis
Market-basket Analysis
Apriori Algorithm
Reccomendation Systems
Collaborative Filtering o User-based o Item-based
Cold-Start Problems
Decision Trees
Classification Inroduction
Hunt’s Algorithm
Tree Induction
Determining the best split
Measures of Node impurity o Gini Index o Information Gain o Gain Ratio
Handling Numeric Attributes
Stopping Criteria for Tree Induction
KNN
Process
Observation
Issues with KNN and Instance based Learning
Engineering KNN
Support Vector Machines (SVM)
Maximum Margin Classifiers
Basic Formulation for Linear SVMs
Dealing with Noisy Data
Non-linear decision boundaries
Practical advice on SVMs
Ensembling Techniques
Ensemble Learning - Motivation
Majority Voting
Bagging Classifiers
Random Forest
Boosting & Stacking o ADA boost o Gradient Boosting Machines o Gradient Descent
Architecting ML Solutions
Summary ML
Data Vizualization
Art and Science of Story telling
Histograms
Box-plots
Bar chart
Pi chart
Scatter plots
Artificial Neural Networks
Limitations of Algorithmic Solutions
Networks in the brain
Perceptrons
Loss Functions
L2, L1, and Huber
Binary – Margin, exponential, Log loss and Hinge loss
Multi – Hinge, Softmax
Regularization Functions
Gradient Descent and local minima
Learning Rate
Multi-Layer Perceptron
Back-propagation Algorithms
Limitations of MLP – Vanishing gradients, and Overfitting
Nurturing a deep neural network
Activating Functions – Sigmoid, tanh, ReLU, Leaky ReLU, ELU
Auto-encoding
Wieght Initializing
Batch Normalization
Dropouts
Convoluted Neural Networks
Understanding images
Traditional approaches
Convolutions – 1D & 2D
Interpreting Convolutions as NNs
How convolutions solves issues
Hyper-parameters – Depth, Size, Stride & padding
Pooling
CNN Architectures - LeNet-5, AlexNet, ZFNet, VGGNet, Inception, ResNet
Pre and Post Processing
Transfer Learning
Recurrent Neural Networks
Motivation
Sequential Data
Architecture
Types
Back Propagation Through Time
Applications
LSTM
GRU
Practical Tips