Introduction to Data Science and Artificial Intelligence
Data science lifecycle, AI history, types of data, problem formulation, data ethics, computational thinking and applications.
CBE Nexus · University Learning Resources
A four-year learning pathway connecting statistics, computation, data engineering and intelligent systems.
Data Science combines statistical reasoning, computing and subject knowledge to extract useful evidence from data. Artificial Intelligence develops computational systems that perform tasks involving prediction, perception, language, planning and decision support. This pathway develops the foundations needed to collect, clean, analyse, model, evaluate and communicate data responsibly.
The sixteen chapters are an illustrative course structure, not a claim that all universities use the same curriculum or award the same degree title. Institutions may place topics in different years, use different prerequisites and require additional mathematics or domain-specific modules. Always compare these notes with your university's approved programme and assessment requirements.
Data science lifecycle, AI history, types of data, problem formulation, data ethics, computational thinking and applications.
Python syntax, control flow, functions, data structures, modules, NumPy, pandas, testing, debugging and reproducible notebooks.
Linear algebra, vectors, matrices, eigenvalues, calculus, gradients, optimisation, logarithms and numerical methods.
Probability rules, random variables, distributions, expectation, variance, sampling, estimation, confidence intervals and hypothesis testing.
Data quality, missing values, outliers, transformations, exploratory analysis, visual encoding, dashboards and communication.
Relational modelling, SQL, joins, aggregation, normalization, transactions, indexing, analytical queries and data warehouses.
Linear models, regression assumptions, diagnostics, ANOVA, generalized linear models, Bayesian inference and uncertainty.
Supervised and unsupervised learning, train-validation-test splits, feature engineering, regularisation, model selection and evaluation.
ETL/ELT, distributed storage, batch and stream processing, data lakes, orchestration, quality monitoring and scalable pipelines.
Perceptrons, backpropagation, activation functions, optimisation, CNNs, sequence models, regularisation and model deployment.
Text preprocessing, tokenisation, language models, embeddings, sequence learning, transformers, information retrieval and evaluation.
Image representation, filtering, feature extraction, classification, object detection, segmentation, convolutional networks and metrics.
Markov decision processes, rewards, policies, value functions, dynamic programming, Q-learning and policy evaluation.
Fairness, bias, explainability, privacy protection, security, model documentation, audit, risk management and governance.
Model serving, APIs, containers, pipelines, experiment tracking, monitoring, drift detection, CI/CD and lifecycle management.
Research design, literature review, data collection, reproducibility, evaluation, technical writing, project governance and presentation.
Students may encounter Python, Jupyter, NumPy, pandas, Matplotlib, scikit-learn, SQL, Git, a deep-learning framework and cloud-based data tools. The choice of tools should follow the learning objective and institutional requirements; software names are not substitutes for understanding the underlying mathematics and assumptions.
Publication note: This page is the course roadmap. Individual chapters must be developed and verified separately before being described as complete.
Use these official organisations for curriculum guidance, examinations and further learning information.