This module focuses on essential Python concepts and techniques for data analytics. The module introduces basic Python concepts, such as the Python interpreter, Jupyter Notebook, input/output, and indentation, enabling students to start developing Python programs for data analytics. Students will learn to apply Python scalar types, objects, attributes, methods, and operators to create and manipulate data structures. They will also apply control statements and iterations, such as conditional statements and loops, to control the flow of execution and process data efficiently. The module covers the use of regular and lambda functions to create reusable and modular code. Additionally, students will learn to apply file-handling techniques to read from and write to files, facilitating data persistence and external data processing. By the end of this module, students will have the necessary Python skills to perform data manipulation, analysis, and processing tasks.

This module explores essential data structures in Python, covering both immutable and mutable types and the powerful NumPy library. Students will learn to apply tuples and strings, along with their methods, to store and manipulate fixed data. They will also apply lists, dictionaries, and sets, as well as their respective methods and operations, to handle changeable data effectively. The module introduces NumPy, enabling students to create, manipulate, and perform arithmetic operations on NumPy arrays using built-in functions. By the end of this module, students will have a solid understanding of Python data structures and NumPy, equipping them with the necessary tools for efficient data manipulation and numerical computations in data analytics tasks.

This module focuses on exploratory data analysis (EDA) and visualisation using the Pandas library and Matplotlib in Python. Students will learn to apply Pandas to create, manipulate, and perform operations on Series and DataFrame objects, enabling efficient data analysis and preprocessing. They will conduct EDA to identify patterns, trends, and relationships in the data. Additionally, students will apply Matplotlib to create informative and visually appealing plots to effectively communicate insights derived from EDA. By the end of this module, students will have the skills to perform comprehensive exploratory data analysis and create meaningful visualisations using Python.

This module focuses on data preprocessing techniques essential for preparing data for analysis. Students will learn to apply methods for reading and writing data in text format while identifying and addressing data quality issues. They will handle missing data by filtering out or filling in missing values and applying various data transformation techniques such as removing duplicates, mapping, replacing values, discretisation, outlier detection and filtering, and encoding categorical variables. Additionally, students will apply data aggregation techniques, including grouping, aggregation and combining functions, to summarise and analyse data. By the end of this module, students will have the skills to preprocess and clean datasets effectively, ensuring data quality and readiness for further analysis.

This module focuses on advanced data preprocessing techniques for handling large and complex datasets. Students will learn to apply data reduction techniques, including dimensionality reduction, numerosity reduction, and sampling methods, to reduce the size and complexity of datasets while preserving important information. They will also apply feature selection techniques, such as filter methods, wrapper methods, and embedded methods, to identify and select the most relevant features for data analysis. Additionally, students will explore feature extraction techniques, including Principal Component Analysis (PCA) and Covariance Analysis, to transform and extract new, informative features from the original dataset. By the end of this module, students will have the skills to effectively preprocess and optimise datasets for improved performance and insights in data analysis tasks.

This module focuses on regression analysis, a fundamental technique in predictive modeling and data analysis. Students will learn to apply linear regression techniques, including univariate and multivariate linear models, to analyse and model the relationship between dependent and independent variables in real-world applications. They will also apply model fitting techniques, such as gradient descent, and evaluate regression models using appropriate metrics to select the best-performing model for a given dataset. Additionally, students will explore nonlinear regression techniques, including smoothing methods, regularised models, robust regression, and nonlinear models, to capture and model complex, nonlinear relationships between variables. By the end of this module, students will have the skills to effectively apply regression techniques to solve real-world problems and make data-driven predictions.

This module focuses on classification techniques, specifically rule-based and parameter-based models. Students will learn to apply decision trees to solve binary and multilabel classification problems and evaluate the performance of these models. They will explore decision tree induction algorithms, considering design issues and measures of impurity, and random forests, to build effective and interpretable models. Students will also apply model selection techniques, such as cross-validation, and address overfitting issues to optimise decision tree models and visualise decision boundaries. Additionally, they will learn to apply logistic regression and discriminant analysis, parameter-based models, to solve classification problems and evaluate its performance. By the end of this module, students will have the skills to effectively apply classification techniques to real-world problems and make data-driven predictions.

This module focuses on unsupervised learning techniques for clustering, which aim to discover natural groupings and patterns in data without prior knowledge of class labels. Students will learn to apply partitional clustering techniques, specifically the k-Means algorithm, considering similarity measures, distance matrices, and cluster goodness evaluation. They will also explore hierarchical clustering methods, both bottom-up agglomerative and top-down divisive, to create nested clusters and analyse data at different levels of granularity. Additionally, students will apply cluster validation techniques, including external and internal indices, to assess the quality of clustering results and determine the optimal number of clusters for a given dataset. By the end of this module, students will have the skills to effectively apply clustering techniques to real-world problems and gain insights from unlabeled data.

This module focuses on privacy, fairness, and security of data analytics. Students will learn about the risk assessment and threat modeling in the practical use of data analytics. Privacy-preserving data mechanism for model privacy will be surveyed. The attack strategies and defense mechanisms of model security will be emphasized. Notions of AI fairness and algorithmic bias will be covered at the stages of pre-processing, in-processing, post-processing stages of data analytics. Cost-sensitive classification and machine learning will be discussed to assess model fairness. Model security will be formalized under frameworks of adversarial data mining for game theory based AI with applications in the cyber kill chain for cybersecurity. Adversarial example games will be summarized for specific targets in adversarial capability, ability and goals. An adversarial risk analysis of the game theories and association optimization trade-offs will be presented in the setup of binary classification, multiclass classification, and multilabel classification. Relation between adversarial and robust data mining for classifier design will be motivated with respect to the robustness properties of analytics models satisfied in defense mechanisms such as semi-supervised machine learning, adversarial training and learning, empirical risk minimization, and mistake-bounds frameworks for adversarial classification. By the end of this module, students will have the skills to effectively apply data analytics techniques to real-world problems and gain insights in a safe, secure, and transparent manner.