Machine Learning Lab PCCSL508 Semester 5 KTU CS 2024 Scheme - Dr Binu V P

About Me- Dr Binu V P

Syllabus

Learn Python well before You start(focus on numpy,pandas,matplotlib )- refer blog

Build a strong understanding of the theory before moving on to programming

Recommended Tools and Setup for Lab

Experiments

Regression 

• Explore Californoa Housing Dataset(learn pandas,scikit-learn and matplotlib)
• Simple Linear Regression using Sample Data ( Single variable, Toy Example)
• Simple Linear Regression using Gradient Descent ( Single variable, Toy Example).
• Simple Linear regression using California Housing Dataset(CSV input).
• Simple Linear Regression Using California Housing Dataset (using scikit-learn).
• Simple Linear regression using Gradient Descent on California Housing Dataset.

• Multiple Linear Regression using using Matrix Form ( Toy Dataset).
• Multivariate Linear Regression using Gradient Descent (Toy Dataset)
• Implement linear regression using scikit-learn on  California Housing dataset( Single Variable).
• Implement linear regression using scikit-learn on  California Housing dataset( Multi Variable).

Polynomial Regression

• Polynomial Regression with varying degrees ( toy example).
• Explore Auto MPG Data set ( learn seaborn).
• Implement polynomial regression on the Auto MPG dataset.
• Bias–Variance Tradeoff using Polynomial Regression on Housing Dataset.

L1 and L2 Regularization

• Comparison of Linear, Ridge, and Lasso Regression with Model Evaluation.
• Implement Ridge and Lasso regression on the Diabetes dataset.

Logistic Regression

• Logistic Regression  using NumPy and  a sample dataset.
• Logistic Regression with Evaluation Metrics and Decision Boundary Visualization( sample data).
• Logistic Regression on Pima Indians Diabetes Dataset-Download the pima indians diabetes dataset.
• Logistic Regression with assessments Confusion Matrix, ROC and AUC.

MLE and MAP

• Comparison of MLE and MAP Estimation using Sample Data.
• Comparison of MLE and MAP Estimation - Beta Prior.
• Comparison of MLE and MAP Estimation- Gaussian Prior.
• Parameter Estimation in Logistic Regression using MLE and MAP (sample data).
• Logistic Regression using MLE and MAP (L1 & L2 Regularization) on Breast Cancer Dataset.
• Multinomial Parameter Estimation using MLE and MAP (Dirichlet Prior) on 20Newsgroups. dataset.

Naive Bayes Classifier

• Prediction of Playing Golf using Naive Bayes Classifier ( Toy example).
• Comparison of Multinomial and Bernoulli Naive Bayes on Play Golf Dataset.
• Text Classification using Multinomial vs Bernoulli Naïve Bayes (Download 20 Newsgroups Dataset).

KNN

• Implementation of K-Nearest Neighbors (KNN) Classification Algorithm-binary class ( toy example).
• KNN for Multi-Class Classification with Visualization and Distance-Weighted Voting.
• KNN Classification using Real Dataset with Performance Evaluation (Download Breast Cancer Data set).
• Image Classification using K-Nearest Neighbors (KNN) on Fashion MNIST Dataset.

Decision Tree

• Basics of Decision Tree using ID3 Algorithm (Iterative Dichotomiser3 Algorithm ).
• Implementation of Decision Tree using ID3 Algorithm with Visualization (scikit-learn).
• Customer Segmentation using Decision Tree (ID3) on Online Retail Dataset ( Download dataset).
• Comparison of Logistic Regression and Decision Tree on Adult Income Dataset (Download dataset ).
  
  

SVM

• Implementation of Linear SVM from Scratch using sample data.
• Implementation of linear SVM using Scikit-learn with Visualization (sample data).
• Classification of Non-Linear Data using SVM with RBF Kernel ( sample data).
• Linear SVM Classification on Iris Dataset with Decision Boundary Visualization ( Download iris data set ).
• Performance Comparison of SVM Kernels on Fashion MNIST Dataset.

Neural Network

• Neural Network for OR Classification with Visualization ( single layer NN).
• Multi Layer Neural Network for XOR Classification with Visualization.
• Implementation of XOR Problem using Neural Network in scikit-learn.
• Implementation of XOR Problem using Neural Network in Keras.
• Implementation of Multilayer Feed-Forward Neural Network (MLP) on Wine Quality Dataset using Keras ( Download Wine Quality Dataset).
• Implementation of Multilayer Feed-Forward Neural Network (MLP) on Wine Quality Dataset with Architecture Comparison.
• Comparison of Activation Functions (Sigmoid, ReLU, Tanh) using Neural Network on MNIST Dataset.
• Hyperparameter Tuning of Neural Network on Fashion MNIST Dataset.

Clustering

• Implementation of K-Means Clustering ( toy example-manual method).
• Implementation of K-Means Clustering Using Scikit-Learn.
• Determining Optimal K using Elbow Method.
• Determining Optimal K using Silhouette Method.
• K-Means Clustering on Digits Dataset with Performance Evaluation.
• Agglomerative Clustering and Dendrogram Visualization (Without Library).
• Agglomerative Clustering and Dendrogram Visualization using Scikit-Learn.
• Comparison of K-Means and Agglomerative Clustering on Mall Customers Dataset (Download Mall Customers Dataset).

Resampling Methods

• Classification using Iris Dataset with K-Fold Cross Validation.
• Bootstrapping with Iris Dataset.
• Bootstrapping vs K-Fold Cross Validation (Iris Dataset).

Ensemble  Methods

• Implementation of Bagging using Random Forest  on Titanic Dataset.(Download  Titanic Dataset)
• Implementation of Boosting using AdaBoost on Titanic Dataset.
• Implementation of Boosting using Gradient Boosting on Titanic Dataset
• Implementation of Boosting using XGBoost on Titanic Dataset
• Implementation and Comparison of Bagging and Boosting on Titanic Dataset