Basic Data Science Course

Lesson 1 • Types of Data and Variables

Classifies data by structure, format, and measurement scale. Enables correct method selection in later analytical chapters.

Lesson 2 • What Data Science Is

Defines data science, distinguishes it from related fields, and maps its components. Grounds all subsequent chapters in shared terminology.

Lesson 3 • The Data Science Workflow

Introduces the standard project lifecycle from problem framing to deployment. Provides a mental model students apply throughout the course.

Lesson 1 • Working with NumPy Arrays

Introduces vectorized computation on numerical arrays. Underpins the performance-critical operations used in pandas and scikit-learn.

Lesson 2 • Control Flow and Functions

Teaches conditionals, loops, and reusable function design. Enables students to automate repetitive data tasks efficiently.

Lesson 3 • File I/O and Data Loading

Demonstrates reading and writing CSV, Excel, and JSON files. Prepares students to ingest real-world datasets in all subsequent projects.

Lesson 4 • Python Syntax and Data Types

Covers variables, operators, and built-in types used in data workflows. Forms the syntactic base for all subsequent Python-based work.

Lesson 1 • Descriptive Statistics

Quantifies central tendency, spread, and shape of distributions. Provides the numerical foundation for interpreting any dataset.

Lesson 2 • Data Visualization Fundamentals

Introduces Matplotlib and Seaborn for creating standard chart types. Visualization skills are applied in every remaining chapter.

Lesson 3 • Data Profiling and Quality Assessment

Systematically audits datasets for missing values, duplicates, and anomalies. Directly motivates the data cleaning techniques in the next chapter.

Lesson 1 • Correlation and Covariance

Quantifies linear and rank-based relationships between variables. Directly informs feature selection and multicollinearity diagnosis in regression models.

Lesson 2 • Probability Distributions

Introduces key discrete and continuous distributions used in data science. Distribution knowledge is essential for choosing correct models and loss functions.

Lesson 3 • Probability Essentials

Covers probability rules, conditional probability, and Bayes' theorem. These concepts underpin classification algorithms and uncertainty quantification.

Lesson 1 • Handling Missing Data

Covers deletion, imputation, and indicator strategies for missing values. Correct handling prevents bias in downstream models.

Lesson 2 • Feature Scaling and Transformation

Applies normalisation, standardisation, and mathematical transforms to numeric features. Ensures distance-based and gradient-based algorithms converge correctly.

Lesson 3 • Building a Preprocessing Pipeline

Combines all cleaning steps into a reproducible scikit-learn Pipeline object. Prevents data leakage and streamlines model training in later chapters.

Lesson 4 • Outlier Detection and Treatment

Identifies outliers using statistical and visual methods and applies appropriate remedies. Protects model performance from extreme value distortion.

Lesson 1 • Machine Learning Fundamentals

Defines supervised learning, the bias-variance tradeoff, and the train-test split paradigm. Establishes the conceptual framework for all modeling sections.

Lesson 2 • Classification Algorithms

Implements logistic regression, decision trees, and k-nearest neighbours for categorical targets. Builds the classification toolkit expanded in the ensemble section.

Lesson 3 • Hyperparameter Tuning

Applies grid search, random search, and Bayesian optimisation to maximise model performance. Tuning is the final step before model selection and deployment.

Lesson 4 • Regression Algorithms

Covers linear, polynomial, and regularised regression for continuous target prediction. Regression skills transfer directly to feature importance analysis.

Lesson 1 • Principal Component Analysis

Reduces feature dimensionality by projecting data onto principal components. PCA improves visualisation and reduces noise before supervised modeling.

Lesson 2 • Manifold Learning Techniques

Applies t-SNE and UMAP for nonlinear dimensionality reduction and visualisation. These methods reveal cluster structure invisible to linear methods like PCA.

Lesson 3 • Clustering Algorithms

Covers k-means, hierarchical, and density-based clustering for grouping unlabelled observations. Clustering outputs feed directly into business segmentation use cases.

Lesson 1 • Communicating Data Science Results

Structures findings into executive summaries and visual dashboards for non-technical audiences. Clear communication determines whether insights drive business decisions.

Lesson 2 • Advanced Model Evaluation

Covers ROC-AUC, precision-recall curves, and calibration for thorough model assessment. Rigorous evaluation prevents costly deployment of underperforming models.

Lesson 3 • Monitoring Models in Production

Detects data drift, concept drift, and performance degradation in deployed models. Ongoing monitoring ensures sustained model reliability after launch.

Lesson 4 • Model Interpretability

Applies SHAP values and permutation importance to explain model predictions. Interpretability builds stakeholder trust and satisfies accountability requirements.