TU Wien:Einführung in Semantic Systems VU (Hose)

Daten

Vortragende	Katja Hose• Emanuel Sallinger
ECTS	3,0
Alias	Introduction to Semantic Systems (en)
Ersetzt	Einführung in Semantic Web VU (Anjomshoaa, Tjoa)
Letzte Abhaltung	2024W
Sprache	English
Abkürzung	SemSys
Mattermost	einfuehrung-in-semantic-systems • Register • Mattermost-Infos
Links	tiss:188399

Zuordnungen

Masterstudium Data Science	Modul VAST/FD - Visual Analytics and Semantic Technologies - Foundations (Pflichtfach)
Masterstudium Business Informatics	Modul ISE/COR - Information Systems Engineering Core (Pflichtfach)
Masterstudium Medizinische Informatik	Modul Informationsverarbeitung (Gebundenes Wahlfach)
Masterstudium Software Engineering & Internet Computing	Modul Informationssysteme (Gebundenes Wahlfach)

Inhalt

Einführung in Semantic Systems teaches the basics of ontology based data modelling. It introduces the concept of an ontology, a lent concept from philosophy that is used to model and represent entities and events with their inter-relationships and properties. Furthermore the technologies of the Semantic Web by Tim Berners-Lee and its development over the past 30 years is taught. That includes the triple storage format of entities in RDF, RDFS and OWL. Finally, the lecture extends the taught knowledge by using state-of-the-art software for Knowledge Graph storage and querying using GraphDB and SPARQL.

Ablauf

2024W: Approximately 6 lectures, with a one month gap in November. The lectures are well made, but it can also be learned by studying the slides and using the internet. Attendance is not mandatory and not necessarily required. During the semester 3 assignments of different degree of complexity and duration are to be completed.

2023W: Lectures introducing the course contents, followed by 3 exercises. Attendance in the lectures does not get checked. The lectures definitely help getting familiar with the contents but one can also simply learn with the slides + internet.

Benötigte/Empfehlenswerte Vorkenntnisse

No previous knowledge or lectures needed. Everything is taught from first principles. Some knowledge in programming is helpful as you may need to use Java or Python libraries in some of the assignments.

Vortrag

noch offen

Übungen

2024W: Exercise 1 was modelling an ontology of your choice in Protégé. The time effort varied between students, due to needing to learn new software and understanding the modelling process. Nevertheless, it was not too hard and required a manageable amount of time. Exercise 2 was about writing different SPARQL queries and executing those against a knowledge graph in GraphDB. It again required installing new software and learning a new syntax, but was overall easy to finish. Exercise 3 was the final exercise. Its task was to create a knowledge graph from existing CSV and JSON datasets. This assignment was considered agonizing by a lot of students. The required software CLI tools were buggy and hard to understand, learn and use, thus blurring any sense of progress. It sparked a long discussion in a TUWEL forum post about its arguably unnecessary complexity. Fortunately, the LVA professors and team seemed understanding and they may adjust the assignment for future semesters.

2023W: The exercises were a bit frustrating because you would spend more time figuring out tools and troubleshooting than actually modelling things. They also differ in difficulty - it can be wise to be strategic and drop some points for some exercise parts. As always, check the minimum points you need per exercise etc.

Prüfung, Benotung

Mix of multiple choice and open questions about understanding. If you did the exercises, the practical part is only of medium difficulty (SPARQL, RDF). The multiple choice questions are of fair difficulty as well.

Dauer der Zeugnisausstellung

2024W: I am writing this as of mid April 2025 and the grades are still not finalized or published. This is due to the second exam results being released together with the one of the first one. Nevertheless, the second exam was mid March, so even in that case we are talking about approximately one month.

The reason for the long time is probably the surprising amount of new students since 2024W. Previous years apparently usually had less new students.

Zeitaufwand

2024W: The classes are fair, but the assignments can be intensively time consuming. Depends on how fast and techy you are to work with buggy tools in assignment 3.

Unterlagen

noch offen

Tipps

The student written course book, which you can find in the Material section, is excellent! It summarizes and sorts the contents of the whole class of 2024 and provides helpful examples for many of the learned concepts. It serves as a great resource to study for the exam.

You should also consider completing the assignments, they prepare well for the practical non theory questions in the exam.

Highlights / Lob

2024W: The exam is great. It's of reasonable complexity and difficulty.

Verbesserungsvorschläge / Kritik

2024W: Assignment 3 was too complex and just overshot any goal of challenging yet successful teaching. Instead it unfortunately only lead to frustration. It took more than 10 hours only on task 1 out of 4, mostly due to buggy tooling and dealing with unknown programs. In its current state it cannot be recommended. The students have provided feedback to the professors and the LVA team, who seemed understanding. They may adapt that assignment in the future.

Materialien

Neues Material hinzufügen

C

• Course Handbook - Tiessler (details)