TU Wien:Description Logics and Ontologies VU (Ortiz de la Fuente)

Aus VoWi
Zur Navigation springen Zur Suche springen

Daten[Bearbeiten | Quelltext bearbeiten]

Vortragende Maria Magdalena Ortiz de la Fuente
ECTS 3
Letzte Abhaltung 2023W
Sprache English
Mattermost description-logics-and-ontologiesRegisterMattermost-Infos
Links tiss:184772, eLearning, Homepage
Zuordnungen
Masterstudium Logic and Computation Wahlmodul Knowledge Representation and Artificial Intelligence


Inhalt[Bearbeiten | Quelltext bearbeiten]

The lecture is based on the book "An Introduction to Description Logic" by Baader et al. The following chapters/topics are mandatory:

  • ALC
  • Expressive DL and Tableau
  • Complexity
  • Lightweight DLs (EL and DL-Lite/QA)

Students can pick one of these two topics and do the exercise for them:

  • Logic Foundations (Model Theory)
  • OWL

Ablauf[Bearbeiten | Quelltext bearbeiten]

Almost weekly lectures (asynchronous on LectureTube) and exercise sheets. Do not underestimate the exercises, they can be quite a lot of work. Exercises are assessed on a scale "not assessable-progressing-mastered". There is no deadline for the exercises other than the final deadline at the end of June. This also means that exercises can be handed in multiple times to achieve a better exercise grade.

There is a one-hour (Zoom) discussion of the material and exercise of the previous week and a look ahead on the week's material. Attendance is not mandatory, but recommended since Prof. Ortiz' explanations are quite helpful for the exercises and exam/quizzes.

There are two quizzes in TUWEL. They have a one-hour time limit, but can be done much faster if you're familiar with the topics. Most questions focus on understanding of the material, so it helps if you've read the chapters in the book and/or watched the lectures attentively.

At the end of the semester, students have to select a topic that was not covered in the lecture and write a short paper (2-3 pages with references), such as:

  • Justification
  • Module extraction
  • Forgetting
  • Repairs
  • ABox abduction
  • Least common subsumer and most specific concept
  • First-order rewritability (of instance queries)
  • DLs with concrete domains
  • DLs with closed predicates
  • Bayesian DLs
  • Defeasible DLs
  • DLs with circumscription
  • Temporal DLs
  • Fuzzy DLs

Benötigte/Empfehlenswerte Vorkenntnisse[Bearbeiten | Quelltext bearbeiten]

Relational calculus, First-Order Logic, Complexity Theory basics as taught in e.g. TIL, DBS, ADBS, FMI, EWBS, KBS, DMOD etc.

Everything else is taught from the ground up, additional material is available if you're not familiar with the topics above.

Compared to Semantic Web Technologies, the material in this lecture is highly theoretical. If you do not like theoretical computer science and logic, you will not have a good time.

Vortrag[Bearbeiten | Quelltext bearbeiten]

The lectures are quite good and cover most of the topics in the book. There are no slides for some of the topics, just "blackboard notes", so it's a good idea to take notes during the lectures. The weekly discussions are there to help with the understanding and the exercises, and often solutions to (sub-) problems were presented.

Übungen[Bearbeiten | Quelltext bearbeiten]

Can be a lot of work, but there is a lot of time to work on them as the only deadline is the end of the semester. However, it's a good idea to hand in early so you can correct any mistakes. It's recommended to write in Latex, but you can hand in any way you want as long as it's legible.

Prüfung, Benotung[Bearbeiten | Quelltext bearbeiten]

Two quizzes, five exercises, one short paper/report.

60% exercise solutions 30% online quizzes 10% reading assignments/special topics

Exercises will be evaluated continuously, you can submit any time

The assessment is based on your mastery of the learning outcomes listed on the sheet:

  • Grade 1 : mastered 100% - 88,00% of the learning outcomes
  • Grade 2: mastered 87,99% - 75% of the learning outcomes
  • Grade 3: mastered 74,99% - 63% of the learning outcomes
  • Grade 4: mastered 62,99% - 50%

Three progressing learning outcomes count as one mastered

Dauer der Zeugnisausstellung[Bearbeiten | Quelltext bearbeiten]

Semester Letzte Leistung Zeugnis
SS21 24.06.2021 29.06.2021 < 1 week

Zeitaufwand[Bearbeiten | Quelltext bearbeiten]

About 10-12 hours per topic for the lecture and exercise. About one day for the report, depending on how fast you can find a suitable survey on the topic.

Unterlagen[Bearbeiten | Quelltext bearbeiten]

An Introduction to Description Logic: Franz Baader, Ian Horrocks, Carsten Lutz, Uli Sattler. Cambridge University Press. (Free access via TU Library)

Tipps[Bearbeiten | Quelltext bearbeiten]

noch offen

Highlights / Lob[Bearbeiten | Quelltext bearbeiten]

noch offen

Verbesserungsvorschläge / Kritik[Bearbeiten | Quelltext bearbeiten]

noch offen

Materialien

Diese Seite hat noch keine Anhänge, du kannst aber neue hinzufügen.