Regression
• Introduction to Linear Regression
• Understanding How Linear Regression Works
• Maths behind Linear Regression
• Ordinary Least Square
• Gradient Descent
• R - square
• Adjusted R-square
• Polynomial Regression
• Multiple Regression
• Performance Measures - MSE, RMSE, MAE
• Assumption of Linear Regression
• Ridge and Lasso regression
• RFE (Recursive Feature elimination)
Hands On - Problem formulation and Case Study on Hotstar, Netflix, And housing prices Dataset
Classification
Logistic regression
• Introduction to classification problems
• Introduction to logistic regression
• Why the name regression ?
• The sigmoid function
• Log odds
• Cost function
• Feature importance and model interpretability
• Collinearity of features
• Feature engineering for non-linearly separable data
Performance Metrics for Classification Algorithms
• Accuracy Score
• Confusion Matrix
• TPR, FPR, FNR, TNR
• Precision - Recall
• F1-Score
• ROC Curve and AUC
• Log Loss
Hands On - Real World Case Study on IBM HR Employee Attrition dataset
K Nearest Neighbors
• Introduction to KNN
• Effectiveness of KNN
• Distance Metrics
• Accuracy of KNN
• Effect of outlier on KNN
• Finding the k Value
• KNN on regression
• Where not to use KNN
Hands On - Different case study on KNN
Natural Language Processing
• Introduction to NLP
• Converting Text to vector
• Data Cleaning
• Preprocessing Text Data - Stop word removal, Stemming , Tokenization, Lemmatization
• Collecting Data from the web
• Developing a Classifier
• Building Pipelines for NLP projects
• Uni-grams,bi-grams and n-grams
• tf-idf
• Word2Vec
Hands On - Text Summarization, WebScraping for data, Sentiment Analysis, Topic Modelling, Text Summarization and Text Generation
Naive Bayes
• Refresher on conditional Probability
• Bayes Theorem
• Examples on Bayes theorem
• Exercise problems on Naive Bayes
• Naive Bayes Algorithm
• Assumptions of Naive Bayes Algorithm
• Laplace Smoothing
• Naive Bayes for Multiclass classification
• Handling numeric features using Naive Bayes
• Measuring performance of Naive Bayes
Hands On - Working on spam detection and Amazon Food Review dataset
Support Vector Machines
• Introduction to SVM
• What are hyperplanes ?
• Geometric intuition
• Maths behind svm’
• Loss Function
• Kernel trick
• Polynomial kernel, rbf and linear kernels
• SVM Regression
• Tuning the parameter
• GridSearch and RandomizedSearch
• SVM Regression
Hands On - Case Study SVM on Social network ADs and Gender recognition from voice dataset
Decision Tree
• Introduction to Decision Tree
• Homogeneity and Entropy
• Gini Index
• Information Gain
• Advantages of Decision Tree
• Preventing Overfitting
• Advantages And Disadvantages
• Plotting Decision Trees
• Plotting feature importance
• Regression using Decision Trees
Hands-On - Decision Tree on US Adult income dataset
Ensemble Learning
• Introduction to Ensemble Learning
• Bagging (Bootstrap Aggregation)
• Constructing random forests
• Runtime
• Case study on Bagging
• Tuning hyperparameters of random forest(GridSearch, RandomizedSearch)
• Measuring model performance
• Boosting
• Gradient Boosting
• Adaboost and XGBoost
• Case study on boosting trees
• Hyperparameter tuning
• Evaluating performance
• Stacking Models
Hands-On - Talking Data Ad Tracking Fraud Detection case study
