Kategorie: ‘Courses’
Bulding Large Language Model Applications – Lab
Large Language Models (LLMs), such as GPT, Claude and Llama, are powerful tools that have transformed the landscape of Natural Language Processing (NLP), enabling advanced applications in various fields. The “Building Large Language Model Applications,” course is a practical, hands-on course designed to provide students with in-depth knowledge and experience in developing applications utilizing LLMs. Throughout the course, students will work on real-world projects and learn how to design, implement, and deploy advanced LLMs systems.
Data Visualisation and Analytics
This course provides participants with a comprehensive and versatile toolbox of data visualisation and analysis methods, which can be transferred to a vast number of applications.
Empathic Computing Lab
Empathic computing investigates how technology can support collaboration, shared understanding and human-centered interaction. In this lab, student teams design and develop their own projects using immersive and pervasive media—ranging from XR headsets and smartphones to tablets, laptops, and wearable devices. Projects may leverage open standards like WebXR, or use engines such as Unity, depending on the goals and platforms chosen. The focus is on creating meaningful, responsive experiences through thoughtful interaction design and emerging technologies.
Knowledge Graphs Lab
The Knowledge Graphs Lab offers a practical insight into structured semantic graphs, which model real-world entities and their complex relationships. By leveraging Knowledge Graphs (KGs), you can represent, integrate, and reason over heterogeneous information in a way that makes data more Findable, Accessible, Interoperable, and Reusable (FAIR).
Why Knowledge Graphs?
Knowledge Graphs power a wide range of applications—from enhancing search engine results (e.g., Google) and fuelling intelligent assistants (like Siri or Alexa) to driving recommendation systems and providing verifiable data backbones for Large Language Models (LLMs)—as they have proven to be scalable, flexible and extendable for storing heterogeneous knowledge across diverse domains.
What You Will Do
- Work in small groups to tackle real-world challenges.
- Design and implement software solutions – from semantically integrating diverse datasets into KGs to integrating KG data into application pipelines.
- Explore how KGs can improve downstream AI applications, such as enhancing the output of LLMs using approaches like GraphRAG
What You Will Gain
- Hands-on experience with popular frameworks for Knowledge Graphs and LLMs.
- Practical insights into building and leveraging KGs, including data modelling, query processing, and semantic integration.
- Teamwork and software development skills
By the end of this lab, you will have a deeper understanding of Knowledge Graph concepts, tools, and how the knowledge of KG can be integrated into real-world applications.
Intelligent Data Management with Generative AI
Generative AI unlocks new potentials supporting tasks in many fields – including data management. It can significantly enhance data management by automating and improving tasks across the data lifecycle. In research data management, it supports metadata creation, documentation, and data curation. For database modeling, it enables schema generation from natural language, query formulation, and schema evolution. In data integration, it assists with semantic mapping and knowledge graph construction. It also can contribute to data quality and governance through supporting anomaly detection and policy drafting and may enhance data ecosystems by aiding cataloging and data exchange. However, the integration of generative AI into data management also presents challenges—ranging from resource efficiency and result quality to issues like bias, hallucination, and data privacy. In this seminar, you will explore a specific subtopic within this field, deepening your understanding while developing your academic writing and presentation skills.
Knowledge Graphs Seminar
Knowledge Graphs are large graphs used to capture information about the real world in such a way that is is useful for applications. In these data structures, there are all sorts of entities (for example, people, events, places, organizations, etc.). Knowledge Graphs are used by many organizations to represent the information they need for their operations. The most well-known example is Google, where a knowledge graph is used to enrich the search results. Also personal assistants, such as Amazon’s Alexa, Apple’s Siri and Google Now, as well as question answering systems such as IBM Watson, make use of knowledge graphs to provide information to their users.
Besides these, also other information graphs, are in use by large organizations to improve or personalize their services. Examples include the Facebook graph, the Amazon product graph, and the Thompson Reuters Knowledge Graph.
The graph also contains all sorts of information about these entities (e.g., age, opening hours, …) and relations between them (e.g., “this shop is located in Aachen”). Furthermore, it may contain context information (e.g., the source of some information) and schema information or background knowledge (e.g., “shops have opening hours”).
Deliverables of this seminar
This seminar consists of an introductory course on Knowledge Graphs. You will give a short outline presentation on your assigned topic to set overview and expectations about the paper you’re going to write. The main deliverable of the seminar is a paper that describes the state of the art of your assigned topic. While you do not need to contribute original research, your task is to show the scientific competences of literature research, presentation of a research question and understanding and putting relevant papers into context. Furthermore, you are asked to critically assess and compare strengths or challenges of existing solutions. You will review your peer’s papers and give relevant feedback to enhance your scientific writing skills. You will present your paper in a final presentation in a block seminar at the end of the semester.
Mixed Reality Lab
Mixed Reality is a continuum of spatial computing experiences on virtual, augmented and extended reality devices, such as the Microsoft HoloLens 2, the HTC Vive Pro, Meta Quest 3 and Android smartphones. In this lab, we learn the basics of mixed reality software development in independent project work that student groups can propose and elaborate.
Privacy Enhancing Technologies for Data Science
This lecture covers current research results in the area of Privacy Enhancing Technologies (PETs) which can be applied to Data Science. These PETs have the potential to enable a new generation of privacy-enabled services which are not focused on maximizing the collection of user data. We use a mix of recent book chapters and papers from conferences and journals of the last few years as primary source material.
Bridge Course Databases
A blended learning bridge course for master students in Data Science, Computational Social Science, and related programs.