Advanced Analytics

Module 1: Advanced ML Topics Using R & Python

Boosting & Bagging

 Intro

 What is Bagging and Boosting

 Comparing the results of Boosting and a single model

 Parameters in Boosting

Gradient Descent

Intro

Concepts of Gradient Descent

Cost function

Learning rate

Extreme Gradient Boosting (XGBM)

Intro

Concepts of XGBM

Parameters in XGBM

Implementation of XGBM

 C5.0

Intro

Concepts of C5.0

Entropy

Information Gain

Forward Pruning

Backward Pruning

Implementation of C5.0

Bias & Variance

Regularization

Module 2: Deep Learning – Neural Network Using R & Python

Deep feedforward networks or Multilayer Perceptrons

Intro

Neurons

Neuron Weights

Activation Function

Networks of Neurons

Input or Visible Layers

Hidden Units

Output Layer

Architecture Design

Gradient-Based Learning

Performance of Deep Learning Models

Empirically Evaluate Network Configurations

Data Splitting

Use an Automatic Verification Dataset

Use a Manual Verification Dataset

Manual k-Fold Cross-Validation

Advanced Multilayer Perceptron

Image Processing models: Convolutional Networks

Convolutional Layers

Filters

Feature Maps

Pooling Layers

Downsampling

Fully Connected Layers

Sequence Modeling: Recurrent and recursive networks

Long Short-Term Memory (LSTM) Networks

Time Series Prediction with Multilayer Perceptrons

Time Series Prediction with LSTM

Recurrent Neural Networks

Module 3: Data Optimization Using XLMiner

Maths behind Optimization

Introduction to derivatives

Derivatives in optimization – Maxima & Minima

Application of optimization in arriving at Linear Least Squares

Gradient Descent Optimization

Linear Programming

Introduction to Linear programming

Formulating linear programming models

Solving linear programming models

Understand resource allocation problems

Understand cost-benefit analysis problems

Duality & other analysis

Decision variables, constraints & objective function

Duality problems

Sensitivity analysis

Network Analysis

Transportation, Shortest path, Maximal flow problems

Introduction to integer linear programming

Introduction to Non-linear optimization

Module 4: Data Simulation Using @Risk

Introduction to Probability

Review of probability

Conditional Probability

Bayes theorem

Permutations & Combinations

 Introduction to Probability Distributions

Bernoulli

Binomial

Geometric

Negative Binomial

Poisson

Uniform Distribution

Triangular

Exponential

Normal

Introduction to Simulation

Basics of simulation

Statistical sampling

The case study on the application of simulation

 Bidding

Marketing

Fitting distributions to data

Decision Tree Simulation

Discrete Event Simulation

Queuing Theory

Module 5: Design Of Experiments Using Minitab

Introduction to DOE

Introduction of DOE terms

Factor, Level, Treatment, Treatment combination

Blocking, Center points, Repetition, Replication

Main effects, Interaction effects

Types of experiments

Trial & Error

One-Factor-At-A-Time (OFAT)

Full factorial design

Fractional factorial design

 Phases of DOE

Screening

Characterization

The 7-step process

Balanced DOE

Calculation of main & interaction effects

Creation of designed experiments

Power & Sample size

Blocking

Defining a custom design

Checking model assumptions

Full factorial results analysis

DOE model reduction

DOE main effect & interaction effect plots

Cube plot, Contour & surface plots

Fractional factorial design

Confounding

Folding

Randomized blocks & Latin square

Implementation plan

Module 6: Natural Language Processing Using R & Python

Introduction to Text Mining & NLP

Factorizing Data

Introduction to topic models

Latent topic modeling

Introduction to parts-of-speech tagging

Perceptual map/bi-plot

Trend tracking – topics across time

Sentence & Word annotations

Named entity annotations

Content Analysis

Lexicons

Emotion Mining – Arcs & emotions

Use of machine learning in text classification

Module 7: Survival Analysis Using R & Python

Introduction to survival analysis

Time-to-event data

Censoring & types of censoring

Survival Analysis Techniques

Single group (Nonparametric methods)

      Life Table

      Kaplan-Meier

      Nelson-Aalen cumulative hazard estimation

Comparison of groups

      Log-rank test

      Wilcoxon test

Semi-parametric estimation mode

Cox proportional hazard model

Survivor function & Hazard function

Bathtub curve

Comparison of survival curves

Failure time distributions

Weibull

Gompertz

Log-logistic

Accelerated event-time

Customer lifetime value

Module 8: Data Analytics Using Spark

Installing & setting up Spark locally

Spark programming in Python

Designing a machine learning system

Obtaining, processing & preparing data with Spark

Building a recommendation engine with Spark

Building a classification model with Spark

Building a regression model with Spark

Building a clustering model with Spark

Dimensionality reduction with Spark

Advanced text processing with Spark

Real-time machine learning with Spark streaming