Course Content
Section outline
-
Session 1: Introduction to Machine Learning (2 Hours)
- Types of ML (Supervised, Unsupervised, Reinforcement)
- Machine Learning Workflow (Data Collection → Preprocessing → Model Selection → Training → Evaluation)
Session 2: Python for Machine Learning (2 Hours)
- Data Manipulation (NumPy & Pandas)
- Data Visualization (Matplotlib & Seaborn)
- Implementing basic ML algorithms using Scikit-learn
Lab Session: Data preprocessing & visualization using real datasets.
Session 3: Supervised Learning – Regression (3 Hours)
- Linear Regression (Simple & Multiple)
- Model Evaluation Metrics (R², RMSE, MAE)
- Regularization (Ridge, Lasso), Polynomial Regression
Lab Session: Predicting stock prices using regression models.
Session 4: Supervised Learning – Classification (3 Hours)
- Logistic Regression, k-NN, SVM, Decision Trees, Random Forests
- Evaluation Metrics (Confusion Matrix, Accuracy, Precision, Recall, F1 Score, ROC Curves)
Lab Session: Train & test classification models on real-world datasets.
-
Session 5: Unsupervised Learning (2 Hours)
- Clustering Algorithms (K-Means, DBSCAN, Agglomerative Clustering)
- Dimensionality Reduction (PCA, t-SNE)
Lab Session: Clustering customer data for market segmentation.
Session 6: Model Evaluation & Hyperparameter Tuning (2 Hours)
- Cross-Validation & Bias-Variance Tradeoff
- Hyperparameter Tuning (Grid Search, Random Search)
Lab Session: Optimizing ML models for higher accuracy.
Session 7: Neural Networks & Deep Learning (3 Hours)
- Introduction to Neural Networks (Perceptrons, Activation Functions)
- Building Deep Learning Models using TensorFlow/Keras
- Applications: CNNs for Image Recognition, RNNs for Time-Series
Lab Session: Implementing a basic neural network.
Session 8: Advanced Topics & Case Studies (2 Hours)
- Reinforcement Learning (Q-Learning, Policy Gradient)
- Real-world case studies (Finance, Healthcare, Retail, E-commerce)
