AI & Machine Learning Using Python

Introduce Artificial Intelligence and Machine Learning, and establish a strong Python foundation for AI work.

Module 1: AI & ML Foundations

• 1.1 introduction to Artificial Intelligence and Machine Learning, AI vs Machine Learning vs Deep Learning, real-world applications of AI

Module 2: Python for AI

• 2.1 Python development environment, variables, data types, operators, conditionals, loops, functions, lists, tuples, sets, dictionaries, strings and data manipulation, list/dictionary comprehensions, modules and packages, exception handling, file handling

Module 3: Professional Tooling

• 3.1 object-oriented programming fundamentals, virtual environments, package management with pip, Jupyter Notebook, Git and GitHub, introduction to NumPy

Capstone: AI Personal Assistant (Python Prototype)

• 4.1 Learners create a Python-based assistant capable of processing user input and performing predefined tasks such as calculations, information retrieval, and simple automation.

Build the mathematical and statistical foundations required for machine learning, and practise data analysis in Python.

Module 1: Mathematics for Machine Learning

• 1.1 variables and functions, linear algebra fundamentals, vectors and matrices, matrix operations

Module 2: Probability & Statistics

• 2.1 introduction to probability, probability distributions, mean, median, mode, variance, standard deviation, correlation and covariance, descriptive statistics, inferential statistics fundamentals, data types and data quality

Module 3: NumPy, Pandas & EDA

• 3.1 NumPy, Pandas, DataFrames and Series, data cleaning: missing values, duplicate records, outlier detection, data transformation, Matplotlib, Seaborn, exploratory data analysis (EDA)

Capstone: Business Data Intelligence & EDA Project

• 4.1 Learners analyse a real-world dataset, clean the data, identify patterns and trends, create visualisations, and produce an analytical report explaining their findings.

Introduce supervised and unsupervised machine learning and the standard model training workflow.

Module 1: Introduction to Machine Learning

• 1.1 supervised, unsupervised, and semi-supervised learning, reinforcement learning introduction, features and target variables, training and testing datasets

Module 2: Preprocessing

• 2.1 data preprocessing, feature scaling, encoding categorical variables, train/validation/test splits, cross-validation

Module 3: Regression & Classification

• 3.1 linear, multiple, and polynomial regression, logistic regression, K-Nearest Neighbours, decision trees, random forests, model training and evaluation, overfitting and underfitting, bias and variance, Scikit-learn

Capstone: Customer Churn Prediction System

• 4.1 Learners develop a machine learning system that predicts whether a customer is likely to leave a service, including data preprocessing, feature engineering, model training, model comparison, evaluation, prediction, visualisation, and business recommendations.

Apply advanced modelling techniques, tuning, and evaluation to real predictive business problems.

Module 1: Advanced Models

• 1.1 advanced regression and classification, Support Vector Machines, Random Forest, Gradient Boosting, XGBoost fundamentals, ensemble learning

Module 2: Tuning & Feature Engineering

• 2.1 hyperparameter tuning: Grid Search, Random Search, cross-validation, feature selection and engineering, dimensionality reduction, Principal Component Analysis (PCA)

Module 3: Evaluation

• 3.1 model interpretability, feature importance, confusion matrix, precision, recall, F1-score, ROC-AUC, regression metrics, model selection, imbalanced datasets, data leakage, machine learning pipelines

Capstone: Predictive Business Analytics Platform

• 4.1 Learners build a predictive system forecasting or classifying a business problem such as sales prediction, customer churn, credit-risk classification, demand forecasting, fraud detection, or property price prediction, comparing multiple models and justifying the final model based on documented evaluation metrics.

Apply unsupervised learning and build recommendation and forecasting systems.

Module 1: Unsupervised Machine Learning

• 1.1 clustering: K-Means, hierarchical clustering, DBSCAN, cluster evaluation, dimensionality reduction (PCA), anomaly detection, association rules

Module 2: Recommendation Systems

• 2.1 collaborative filtering, content-based recommendations, similarity measures, search and ranking concepts

Module 3: Time-Series & Automation

• 3.1 time-series fundamentals: trend and seasonality, forecasting fundamentals, AI automation, model pipelines

Capstone: AI Recommendation & Customer Intelligence System

• 4.1 Learners build an AI system that analyses customer behaviour and generates personalised recommendations, with possible applications in product, food, course, property, or content recommendation, or customer segmentation.

Introduce neural networks and deep learning, and build a computer vision application.

Module 1: Neural Network Fundamentals

• 1.1 biological vs artificial neurons, perceptrons, neural network architecture, activation functions, forward propagation, backpropagation, loss functions

Module 2: Training & Optimisation

• 2.1 optimisation, gradient descent, learning rates, epochs and batches, regularisation, dropout, batch normalisation

Module 3: Frameworks & CNNs

• 3.1 TensorFlow fundamentals, Keras, PyTorch introduction, Convolutional Neural Networks, image classification, transfer learning, model evaluation, GPU acceleration

Capstone: AI Image Classification System

• 4.1 Learners develop a deep-learning application capable of classifying images (e.g. medical images, animals, products, plant disease, documents, or general object recognition), including model training, validation, testing, and an application interface for making predictions.

Apply NLP and generative AI techniques, and build a retrieval-augmented AI assistant.

Module 1: NLP Fundamentals

• 1.1 text preprocessing: tokenisation, stop words, stemming and lemmatisation, Bag-of-Words, TF-IDF, text classification, sentiment analysis, Named Entity Recognition

Module 2: Transformers & LLMs

• 2.1 word embeddings, transformers, attention mechanisms, Large Language Models, generative AI fundamentals, prompt engineering, AI APIs, open-source models, the Hugging Face ecosystem

Module 3: RAG & Responsible AI

• 3.1 embeddings, vector databases, semantic search, Retrieval-Augmented Generation (RAG), AI chatbots, AI agents fundamentals, responsible AI, hallucination and model limitations, AI evaluation

Capstone: AI Knowledge Assistant / RAG Chatbot

• 4.1 Learners develop an intelligent chatbot that answers questions from a defined knowledge base, demonstrating document ingestion, text processing, embeddings, vector search, retrieval, LLM integration, context-aware responses, conversation history, source/reference retrieval, and basic safety controls, for example a university, healthcare, business-support, cybersecurity, or customer-service assistant.

Deploy, monitor, and govern AI systems in production, and complete the final capstone.

Module 1: AI System Architecture

• 1.1 the machine learning project lifecycle, AI system architecture, model packaging, model serialisation, REST APIs for AI models: FastAPI, Django AI integration

Module 2: Deployment & Monitoring

• 2.1 Docker fundamentals, Git/GitHub, CI/CD for AI applications, cloud deployment concepts, model serving, batch vs real-time inference, model monitoring, data drift, model drift, model versioning, experiment tracking, ML pipelines, MLOps fundamentals

Module 3: Responsible & Scalable AI

• 3.1 logging and monitoring, AI security, privacy and data protection, responsible AI, bias and fairness, AI governance, cost optimisation, scaling AI applications, technical documentation, AI portfolio development, production deployment

Capstone: Production AI & Machine Learning Platform

• 4.1 The final capstone: learners design and develop a complete AI-powered application solving a genuine business or societal problem, for example an AI healthcare, education, cybersecurity, business intelligence, or food intelligence platform, demonstrating data acquisition, cleaning, and exploratory analysis, machine learning/deep learning, model evaluation, AI integration, API development, database integration, a user interface, authentication/security, model deployment, monitoring, documentation, and Git/GitHub version control.