UMBC CMSC 491/691 Fall 2022
Knowledge Graphs

Home · Schedule · HW · Exams · Notes · GitHub · Examples · Resources · Colab · Discord · Webex

Notes on coverage for the 2018 final exam

The 2018 final exam will be comprehensive.

Make sure you've read chapters 1-4 of the Semantic Web Primer as well as sections 7.1-7.3 and 7.6-7.7

Review the presentation slides we covered in class

Understand the basic concepts underlying

  • Ontologies and their role in providing a semantic schema for data
  • XML and JSON, including their data models and serializations
  • RDF, RDFS, OWL, including the data models and serializations (e.g., N3, Turtle, XML, JSON-LD and Manchester syntax (used in Protege)
  • How a reasoner (like ones built into Protege) can use axioms derived from the meaning of RDFS and OWL constructs to infer new relations and identify contradictions in a knowledge graph
  • The use explicit rules encoded in N3 or SWRL for making inferences from RDF data.
    • Understand the limitations of both OWL description logic reasoning and reasoning with SWRL rules, i.e., being able to give an example of something each can do that the other cannot
    • Be able to express an axiom given in OWL in SWRL and vce versa when it can be encoded both ways
  • Understand the new profiles introduced in OWL2 (QL, RL and EL), the motivation for introducing them, and what limitations they each embody
  • The ecosystem for creating and using semantic web data and knowledge: ontology editors (e.g., Protege), triple stores, SPARQL endpoints, browser-based knowledge graph explorer systems, (e.g., http://dbpedia.org/page/Alan_Turing)
  • SPARQL as a query/update language for RDF knowledge graphs, writing simple SPARQL queries
  • The use of Schema.org as a markup language/ontology supported by companies.
  • The three schemes for adding semantic markup to HTML: Microdata, RDFa and JSON-LD
  • Understand the motivation for Wikidata and how it fits into the Wikimedia community

In addition to questions at a conceptual level, it's likely that there will be the following kinds of questions

  • Given a graphical depiction of an RDF graph, answer simple questions about it or give a serialization in Turtle
  • Given a Turtle serialization, draw the graph of nodes and links that it represents and/or describe what it means in one or more English sentences
  • Given a Turtle serialization of a small graph, answer questions about what it does or doesn't not imply
  • Given a an English description of some entities and facts and some OWL classes and properties, represent the facts in the description
  • Write one or more SWRL rules in the Manchester syntax (i.e., what Protege uses) to capture the meaning of a statement, e.g., if two people share a parent, they are siblings
  • Given a set of triples and a SPARQL query, describe what would be returned by the query
  • Given HTML with embedded markup in Microdata or RDFa, list the triples that it contains

You should also review relevant questions from past midterm and final exams