About the exam
There's a three-hour written exam in INFO216. It counts 60% of the final grade, whereas the group assignment (the programming project) counts 40%. Here is some more information about the planned exam format, so you can spend those three hours more effectively to demonstrate what you know.
The type of questions can vary from more concrete/detailed tasks, theoretical questions/quiz to argumentative/reflective tasks.
Examples of concrete/detailed exam tasks
Here are some examples of concrete/detailed exam tasks you may encounter. Of course the possible tasks are not limited to this. They are only examples.
- given a situation/scenario, draw an RDF graph
- extend or find problems in an RDF graph
- either of the above using RDFS or OWL concepts
- either of the above using concepts from some vocabulary
- given a situation, find OWL-classes, -properties and -individuals
- given an OWL-ontology, make it more precise, e.g., using property restrictions
- given an OWL-ontology, define new composite concepts
- express an RDF, RDFS or OWL graph in Turtle, JSON-LD or another format
- given a SPARQL query or update, describe the result
- write a SPARQL query or update
- extend or find problems in a SPARQL query or update
- write or outline Jena code for creating the above RDF graphs
- write or outline Jena code for creating the above SPARQL queries and updates
- extend or find problems in Jena code like the above
Examples of argumentative/reflective tasks
Here are some examples of questions that will require answers of an argumentative/reflective type:
- What are the main concepts in RDF/RDFS/OWL for connecting or relating vocabularies semantically? Give examples of how each of them can be used, using vocabularies you make up or (better) already know.
- What are the main uses of blank (or anonymous) nodes in RDF/RDFS and in OWL.
- Show the general architecture a semantic application in a drawing, and explain each of the main components briefly.
- Show the general architecture of Jena - at least the packages we have been looking at in the course - and explain briefly the purpose and contents of each package.
- Your department is going to develop a new database for keeping track of all kinds of production-related equipment and assets (production machines, computers, software, transportation vehicles). A lot of the equipment and assets are rented from/to and used in collaboration with business partners, and the types of equipment and assets changes from project to project. You think this database should be a triple store, but the other people in your department are used to relational database. Present your main arguments for using a triple store for this type of system.
- You are brought in as an advisor for a group of small and medium-sized companies that produce feed for fish farms in Western-Norway. They want to start using semantic technologies to make information sharing quicker and easier, but there are no suitable vocabularies available, except for a few general biological ones that describe the the fish and other species involved. How would you advise them to go about developing a new vocabulary?
- What are the main advantages of the web of data (or semantic web) compared to the plain old (non-semantic) web?
- What are the "closed world assumption" (CWA) and the "open world assumption" (OWA)? How are they relevant for the web of data (semantic web)? Which of them are used in the web of data at large? Which constructs in RDF/RDFS are related to the CWA/OWA distinction?
- A company has made a new vocabulary and documented it as an OWL DL ontology. They ask you to evaluate it. Describe how you would go about evaluating the ontology. For example, which criteria would you use and/or which questions would you ask, who will you involve and which actions will you take?
When you answer questions like these, it is important to remember that we do not invite loose opinions, hunches or feelings. What we want are opinions or viewpoints backed by clear arguments, facts and demonstrated knowledge of good practices.