Machine Learning

Topics Covered during Live Sessions

Week 1: Linear Regression & Gradient Descent Algorithm

• What is Artificial Intelligence
• Machine Learning & Types
• Fundamentals of Python
• Data Preprocessing in python
• Defining a Model
• Error Calculation
• Gradient Descent Algorithm

Problem: Predicting Housing Prices based on the size of the house

Week 2: Logistics Regression & Classification

• Classification problem
• Data Preprocessing
• Defining a Model
• One vs. All
• Error & Accuracy Calculation
• Gradient Descent Algorithm
• Prediction

Problem: Handwritten Digit Recognition

Week 3: Decision Tree & Random Forest

• What is Decision Tree?
• Calculating Entropy
• Calculating Information gain
• Gini Index
• Objective Function
• What is Random Forest & Bagging?
• Advantages of using Random Forest
• Pruning
• Objective Function
• Prediction & Accuracy

Decision Tree Problem: Depending on the weather predicting the possibility of playing cricket
Random Forest Problem: Behavioral Risk factor Surveillance System

Week 4: Naive Bayes

• NLP Basics
• N Gram Model, Bag of Words
• TF-IDF Vectorisation
• Bayes Theorem
• Multinomial & Bernoulli Naive Bayes

Problem: Building a spam classifier

Week 4: Boosting Techniques

• What is Gradient Boosting?
• Adaboost | Xgboost 
• Algorithmic Enhancements
• System Optimization
• Objective function
• Prediction & Accuracy

Problem: Indian Pima Diabetics Prediction

Week 5: Support Vector Machine (SVM)

• What is SVM?
• Intuition Behind SVM
• Defining the Objective function
• Optimize Objective function
• Kernels
• Prediction & Accuracy
• HOG SVM
• Sliding Window technique
• Optimisation

Problem: Solving MNIST dataset, Facial Recognition

Self Learning Content

K Nearest Neighbour & SMOTE

• What is KNN?
• Example KNN Problem
• Defining the Objective function
• Optimize Objective function
• Prediction & Accuracy

Problem: Online shoppers’ buying intention

Neural Networks

• What is Neural network?
• Defining a Model
• Back Propagation Algorithm
• Classification Problem

Problem: Solving Fashion MNIST dataset

Unsupervised Learning & K Means Clustering

• K Means Clustering
• Hierarchical Clustering
• K Means for non-separated clusters
• Principle Component Analysis

Problem: Movie Recommendations