AI/ML FOR PRACTITIONERS

Overview

This course will help professionals develop a fundamental understanding of machine learning and derive practical solutions using predictive analytics. It introduces the concepts related to Supervised and Unsupervised machine learning from basic regression and classification to decision trees and clustering. The course will make use of Python for the hands-on implementation of the models.

Learning Objectives

  • Be able to understand and analyze AI/ML opportunities, have balanced conversations with your clients on AI/ML topics.
  • Develop foundation background in Python and tools for Data analytics and ML.
  • Perform data analytics with a graphical representation using Python.
  • Handle data analytics with a graphical development environment, which makes advanced tools easily accessible without coding.
  • Conduct and interpret some basic data science activities.
  • Develop understanding of what is Machine Learning using basic machine learning experiment and how to interpret its output.
  • Understand ML Techniques and their applicability in business via case studies.
  • Learn how computing systems take decisions with minimal human intervention.
  • Learn how AI systems change behaviours without being explicitly programmed.
  • Discuss major drivers for the growth of AI in verticals such as automotive, finance, Insurance, Real estate, Media and advertisement etc.
  • Detailed discussion on major case studies in different verticals such as automotive, finance, Media, Insurance, Sales with deployment approach.

Prerequisites

Technical Background:

We recommend that all students have:

  • A basic understanding matrix vector operations and notation
  • A basic knowledge of statistics
Hardware / PC capabilities:

A workstation with the following capabilities:

  • A web browser (Chrome/Firefox)
  • Internet connection
  • A firewall allowing outgoing connections on TCP ports 80 and 443
  • The following developer utilities should be installed:
    • Anaconda
    • Jupyter Notebook (will be installed using Anaconda)
    • All libraries will be installed using Anaconda
Software Requirements:
  • All code will be written in Python with the use of the following libraries: Pandas/NumPy are the libraries for matrix calculations and data frame operations.
  • We strongly recommend to browse through the available tutorials for these packages, for instance, the official one scikit-learn, Matplotlib
  • If software requirements cannot be satisfied due to the security policy of your employer, please inform us about the situation to find an appropriate solution for this issue.

Course Content

Unit:1 Introduction for Data Analytics, Machine Learning and Artificial Intelligence

  • What is Data Science
  • Applications of Data Science
  • Case Study : Investment research analysis
  • What is AI / ML
  • Typical Customer Segmentation Workflow
  • Case study: Building a customer service chatbot

Unit 2: Machine Learning: Understanding jargons

  • Machine Learning Basics
  • Supervised machine learning
  • When to use Supervised Learning
  • Features and labels
  • Model evaluation
  • Clustering
  • Supervised vs. Unsupervised
  • When to use Unsupervised Learning
  • Cluster size selection

Unit 3: Building a data science team and responsibility assignment

  • AI-ML Project cycle Overview
  • AI-ML Job Descriptions and skill sets
  • Matching skills to jobs
  • Assigning data science projects
  • Interpreting a team sprint
  • Classifying data tasks

Unit 4: Python: Journey from Foundation Level

  • Basics of Python
  • Data types (List, Tuple, Set, Dictionary, Data Frames)
  • Operators (Arithmetic, Assignment, Comparison, Logical, Identity, Membership)
  • Conditions and Nesting
  • Python Classes and Objects
  • Looping mechanisms, IN operator
  • IO Handling, Files I/O
  • Database Access
  • Python Functions, Modules

Unit 5: Advanced Topics Overview in Machine Learning

  • Classifying machine learning tasks
  • Sentiment Analysis (Natural Language Processing)
  • Deep Learning and Explainable AI
  • Where should I use Deep Learning?
  • Finding the correct solution to given problem
  • Google Auto-ML usage to implement all these things very easily for non-developers

Unit 6: Statistical : Foundation building Block for Machine Learning

  • Descriptive Statistics,
  • Laws and Axioms of Probability,
  • Probability Distribution,
  • Hypothesis Testing and Scores
  • Hands-on practice
  • Stochastic Gradient Descent Optimization,
    coefficient of determination,
    significance tests
    Confidence and prediction intervals,
    categorical variables
    Outliers, auto-regression and transformation of variables,
    Polynomial Regression
  • Ensemble Techniques: Bagging and Boosting
  • Random Forests, Feature importance
  • Stacking
  • Hands-on practice in Python

Unit 7: Applied Python with data analytics Libraries

  • Pandas for importing data and pre-processing
  • Numpy for Statistical Analysis
  • Matplotlib for Data Visualization
  • Hands-on practice

Unit 8: Foundation building in Machine Learning Techniques

  • Supervised Machine Learning Algorithms
  • Application in predictive Analytics
  • Linear Regression: Single and Multiple Linear Regression (Estimation)
  • Modelling and Prediction,
  • coefficient of determination,
  • confidence and prediction intervals,
  • categorical variables, outliers
  • Hands-on Demo

Unit 9: Supervised Machine Learning with application in Classification (Prediction)

  • Linear Classification: Logistic Regression
  • Implementation and optimization,
  • Estimation of probability using logistic regression,
  • ROC Curve, Feature selection in logistic regression
  • Naïve Bayes: Bayes Theorem, Naïve Bayes Classifier
  • K Nearest Neighbor Algorithm (KNN)
  • Support Vector Machine: Linear Support Vector Machine, Kernel-based Classification, Controlled Support Vector Machine, Support Vector Regression
  • Decision Tree: Training and Visualizing Decision Tree, CART Training algorithm, Impurity measures, Gini Impurity index, Cross-entropy impurity index, Misclassification impurity index, feature importance in tree
  • Various time series models for modelling and predicting
  • Hands-on demo in Python

Unit 9: Unsupervised Machine Learning: Clustering

  • K Means clustering
  • Hierarchical clustering
  • PCA (Principal Component Analysis) and its Applications
  • Hands-on Demo

Unit 10: Case studies: Discussions and implementations

  • Recommendation of Wines,
  • Employee productivity via ML,

Unit 10: Case studies: Discussions and implementations

  • Healthcare chatbots
  • Automatic Music Tagging
  • Smart Cars.

Unit 11: Deep Learning foundation

  • Fundamentals of Neural Network Analysis (Demo)
  • Introduction to Recommendation System (Demo)
  • Introduction to Sentiment Analysis (Demo)
  • Introduction to Deep learning and TensorFlow (Demo)