| CODE | ICS5111 | ||||||||||||
| TITLE | Mining Large-Scale Data | ||||||||||||
| UM LEVEL | 05 - Postgraduate Modular Diploma or Degree Course | ||||||||||||
| EQF/MQF LEVEL | 7 | ||||||||||||
| ECTS CREDITS | 5 | ||||||||||||
| DEPARTMENT | Artificial Intelligence | ||||||||||||
| DESCRIPTION | This study-unit will present two important and complimentary aspects of how to deal with large-scale data, namely text and graph mining. In the first part of the study-unit the students will be exposed to various techniques that support three important aspects related to large data (structured or unstructured): (i) acquisition of information (ii) insight (identifying patterns and trends upon which to base actions), and (iii) prediction (modelling future activities or outcomes based on input data). The topics that will be tackled include: - Text classification and analytics; - Document clustering and summarisation; - Data cleanup techniques; - Frequent-itemset mining; - Feature selection; - Dimensionality reduction; - Algorithms for analysing and predicting trends from different data sources. In the second part of the study-unit the focus will be on graph mining. Students will become familiar with the graph data structure, the challenges of processing large amounts of data as graphs, state-of- the-art methods and algorithms or analysing graphs, and applications of graph mining in various domains. The topics that will be covered include: - Introduction to Graph and Graph Machine Learning; - Graph Embeddings; - Graph Neural Networks; - Knowledge Graphs; - RAG and Graph-RAG. Study-Unit Aims: - Provide a rigorous introduction to text mining and graph mining for large-scale data, covering the full pipeline: acquisition, insight, prediction across structured and unstructured sources; - Develop students’ ability to model data representations and to select appropriate algorithms for classification, clustering, summarisation, analysis, and link/influence analysis; - Equip students to design, implement, evaluate, and communicate end-to-end data mining solutions using modern open-source tools in Python, with attention to scalability, reproducibility, and evaluation rigor; - Introduce state-of-the-art methods for graphs and for text, preparing students to read and apply current literature; - Leverage a blended delivery model to strengthen independent learning, collaboration, and reflective practice through a mix of short lectures, guided labs, structured asynchronous activities, and mini-projects. Learning Outcomes: 1. Knowledge & Understanding: By the end of the study-unit the student will be able to: - Explain the theoretical foundations of text mining (vector space representations, n-gram/linguistic features, basic summarisation, document clustering, frequent-itemsets, feature selection, dimensionality reduction) and graph mining (graph types, connectivity, small-world/scale-free properties, centrality, community structure); - Compare alternative data representations and their trade-offs for large-scale tasks; - Describe modern graph-based AI approaches based around Graph Neural Networks and outline how Knowledge Graphs and Graph-RAG systems support grounded retrieval and reasoning; - Discuss evaluation metrics and scalability considerations for text and graph algorithms. 2. Skills: By the end of the study-unit the student will be able to: - Design and implement reproducible text- and graph-mining pipelines in Python using open-source libraries; - Analyse large-scale networks and corpora, computing and interpreting connectivity, paths, centrality, community structure, classification/clustering, summarisation, and trend metrics with sound validation; - Select and justify representations and algorithms (e.g., TF-IDF vs embeddings; property graphs vs knowledge graphs; link analysis vs GNNs) with attention to scalability, complexity, and resource constraints; - Build and query small knowledge graphs and integrate retrieval (Graph-RAG) to ground answers/predictions; prototype basic graph embeddings/GNN workflows where appropriate; - Communicate findings clearly to technical and non-technical audiences through concise reports, executable notebooks, short demos, and reproducible artefacts (code, data/model cards, experiment logs); - Collaborate effectively in a blended setting, using version control, peer review, and issue tracking to manage scope, time, and shared deliverables; - Apply responsible AI practice (privacy, data governance, bias/fairness) and reflect on limitations and risks, articulating ethical trade-offs and mitigation steps. |
||||||||||||
| STUDY-UNIT TYPE | Lecture and Online Learning | ||||||||||||
| METHOD OF ASSESSMENT |
|
||||||||||||
| LECTURER/S | Charlie Abela Joel Azzopardi |
||||||||||||
|
The University makes every effort to ensure that the published Courses Plans, Programmes of Study and Study-Unit information are complete and up-to-date at the time of publication. The University reserves the right to make changes in case errors are detected after publication.
The availability of optional units may be subject to timetabling constraints. Units not attracting a sufficient number of registrations may be withdrawn without notice. It should be noted that all the information in the description above applies to study-units available during the academic year 2026/7. It may be subject to change in subsequent years. |
|||||||||||||