Data Science and Data Analytics Course

Lesson 1 • The Analytics Spectrum

Introduces descriptive, diagnostic, predictive, and prescriptive analytics. Students map each type to business questions and expected outputs.

Lesson 2 • What Data Science Is

Defines data science, its scope, and how it differs from statistics and software engineering. Establishes vocabulary used throughout the course.

Lesson 3 • Data Types and Structures

Covers structured, semi-structured, and unstructured data formats. Connects data type awareness to appropriate tool and method selection.

Lesson 1 • NoSQL and Cloud Storage Options

Surveys document, key-value, and columnar NoSQL stores alongside cloud object storage. Students select storage types based on data volume and query patterns.

Lesson 2 • Primary and Secondary Data Sources

Distinguishes first-party, second-party, and third-party data and their trade-offs. Grounds source selection in reliability and relevance criteria.

Lesson 3 • Building Simple Data Pipelines

Combines collection and storage skills into automated ingestion workflows. Students write scripts that extract, transform, and load data reliably.

Lesson 4 • Web Scraping and APIs

Teaches programmatic data retrieval using REST APIs and HTML scraping. Students extract and parse real data from live web sources.

Lesson 1 • Data Transformation Techniques

Applies normalisation, standardisation, encoding, and binning to prepare features. Connects transformations to downstream model compatibility requirements.

Lesson 2 • Reshaping and Merging Datasets

Teaches pivoting, melting, concatenating, and joining multiple DataFrames. Students consolidate multi-source data into unified analytical tables.

Lesson 3 • Exploratory Data Analysis Basics

Uses summary statistics and distributions to understand dataset characteristics. Provides the diagnostic lens needed before any cleaning begins.

Lesson 4 • Handling Missing and Corrupt Data

Covers detection, imputation, and removal strategies for incomplete records. Students choose methods based on missingness mechanisms (MCAR, MAR, MNAR).

Lesson 1 • Principles of Effective Visualisation

Covers perceptual principles, chart selection, and common visualisation pitfalls. Anchors design decisions in audience comprehension and data integrity.

Lesson 2 • Interactive Visualisation with Plotly

Creates hover-enabled, zoomable charts and dashboards using Plotly and Dash. Extends static skills to web-ready, user-driven exploration.

Lesson 3 • Data Storytelling and Presentation

Structures analytical findings into coherent narratives for executive audiences. Students craft slide decks and reports that drive decisions.

Lesson 4 • Geospatial and Multivariate Visuals

Introduces choropleth maps, heatmaps, and parallel coordinates for complex data. Students visualise spatial and high-dimensional relationships effectively.

Lesson 1 • Hypothesis Testing Framework

Introduces null and alternative hypotheses, p-values, and Type I/II errors. Students apply the framework to real business and scientific questions.

Lesson 2 • Common Statistical Tests

Applies t-tests, chi-square, ANOVA, and Mann-Whitney tests to appropriate data types. Students select tests based on data scale and distribution assumptions.

Lesson 3 • Correlation and Regression Analysis

Measures linear relationships and builds simple and multiple regression models. Bridges descriptive statistics to predictive modelling introduced later.

Lesson 4 • Probability and Distributions

Covers probability rules, random variables, and key distributions (normal, binomial, Poisson). Provides the mathematical foundation for all inferential work.

Lesson 1 • Regression Algorithms

Extends linear regression to regularised models (Ridge, Lasso) and tree-based regressors. Students predict continuous outcomes and interpret coefficients.

Lesson 2 • Core ML Concepts and Workflow

Defines training, validation, and test splits, bias-variance trade-off, and overfitting. Establishes the end-to-end ML pipeline used in every subsequent model.

Lesson 3 • Model Evaluation and Selection

Uses cross-validation, confusion matrices, ROC-AUC, and RMSE to assess models. Students select the best model for a given task using objective metrics.

Lesson 4 • Classification Algorithms

Covers logistic regression, decision trees, k-NN, and support vector machines. Students train and compare classifiers on labelled datasets.

Lesson 1 • Hyperparameter Optimisation

Applies grid search, random search, and Bayesian optimisation to tune models. Students automate tuning and document performance gains systematically.

Lesson 2 • Neural Network Foundations

Builds feedforward networks with Keras, covering layers, activations, and backpropagation. Connects neural network mechanics to the deep learning models that follow.

Lesson 3 • Natural Language Processing Basics

Covers tokenisation, TF-IDF, word embeddings, and transformer-based classification. Students build text classifiers and sentiment analysers.

Lesson 4 • Ensemble Methods

Covers bagging, boosting, and stacking with Random Forest, XGBoost, and LightGBM. Students tune ensembles to achieve state-of-the-art tabular performance.

Lesson 1 • Building REST APIs for Models

Wraps trained models in FastAPI endpoints for real-time inference. Students test, document, and secure prediction APIs.

Lesson 2 • Model Packaging and Serialisation

Saves and loads models using pickle, joblib, and ONNX for portability. Ensures reproducibility across environments and team members.

Lesson 3 • Containerisation with Docker

Packages model APIs into Docker containers for consistent, portable deployment. Students write Dockerfiles and manage images for ML services.

Lesson 4 • Cloud Deployment and Scaling

Deploys containerised models to cloud platforms using managed services. Students configure auto-scaling and load balancing for production traffic.