TDDD43 Advanced Data Models and Databases
Examples of exam question types
This is a collection of example exam question types.
Some old exams can be found here. (Observe that some topics in the old exams may not be included anymore, while some new topics may have appeared.)
- Describe the components of IR models.
- Explain and exemplify the basic principles of boolean/vector/probabilistic model for IR (different components).
- Describe the advantages and drawbacks of the boolean/vector/probabilistic model.
- Explain TF-IDF.
Semi structured data, XML, RDF(MAY BE UPDATED)
- Describe general properies of semi-structured data.
- Model a given scenario or set of data in the OEM model.
- Explain a Lorel query.
- Describe the general idea and define the properties of a data guide.
- Define minimal and strong data guides.
- Given a description of data, construct minimal and strong data guides for that data.
- Describe the general properties and concepts of XML.
- Describe the main concepts of the RDF data model.
- Model a given scenario (text or ER) using XML.
- Model a given scenario (text or ER) using RDF.
- Use and understand DTD, XML schema.
- Write an XPath query.
- Write an XQuery query.
- Explain a SPARQL query; show the result of a SPARQL query on given data.
- Explain what Linked (Open) Data is.
Semantic web and ontologies
- Describe the problems in the current web and the vision of the semantic web.
- Show how semantic annotations based on ontologies would help solve the problems of the current web.
- Explain what ontologies are used for.
- Describe and explain the OBO Foundry principles.
- Explain and give examples of the components of ontologies.
- Describe from a knowledge representation point of view the different kinds of ontologies.
Description logics and OWL
- Describe the notions of T-box, A-box, knowledge base, subsumption, satisfiability.
- Model a given a scenario using description logics.
- Give the semantics for a given description logic construct.
- Describe the difference between open-world assumption and closed-world assumption.
- Describe the different reasoning services.
- Given two concepts, prove that one concept subsumes the other (or not) using a tableau algorithm.
- Know some reasons for intractibility, undecidability of description logics.
- Know the difference between the variants of OWL.
NoSQL concepts, techniques and databases(MAY BE UPDATED)
- Explain the main reasons for why NoSQL data stores appeared.
- List and describe the main characteristics of NoSQL data stores.
- Discuss the trade-off between consistency and availability in a distribute data store setting.
- Discuss different consistency models and why they are needed.
- Explain the CAP theorem; Explain the difference between ACID and BASE properties.
- Explain how consistency between replicas is achieved in a distributed data store.
- Explain the differences between vertical and horizontal scalability.
- Explain how consistent hashing works and what problems it addresses.
- Explain how vector clocks work and what problems they address.
- Classify NoSQL data stores according to their data model and provide example applications for the different models.
- Explain what HDFS is and for what types of applications it is (not) good for.
- Explain the organization of HDFS.
- Explain the basic principles and steps of MapReduce.
- Give examples of applications that benefit from the MapReduce model.
- Model a given scenario (text or ER) using a Property Graph.
- Explain a Gremlin expression; show the result of a Gremlin expression on a given Property Graph.
- Explain a Cypher query; show the result of a Cypher query on a given Property Graph.
- Describe the drawbacks of using MapReduce for iterative graph algorithms.
- Name the steps of a vertex compute function as used in vertex-centric graph processing systems.
Ontology alignment and debugging
- Describe the problem of ontology alignment.
- Describe a framework for ontology alignment and explain the different components.
- Explain and give examples for different strategies for preprocessing/matching/combining/filtering.
- Explain how to evaluate ontology alignment strategies / systems.
- Understand performance of different approaches.
- Know challenges for ontology alignment.
- Explain and exemplify different kinds of defects in ontologies.
- Give examples of the influence of defects in ontologies for semantically-enabled applications.
- Describe a framework for ontology debugging and explain the different components.
- Explain/compute MIPS and MUPS.
- Explain the formulation of GTAP.
- Expain the intuitions behind the preferences (semantically maximal, subset minimal, minmax, maxmin, skyline optimal) for GTAP solutions.
Data source integration
- Describe the problems occuring with data source integration.
- Describe the different steps in data source integration.
- Describe the method for integration based on link driven federations.
- Understand a query in the SRS query language.
- Know advantages and disadvantages of link driven federations.
- Describe the method for integration based on mediation.
- Describe mappings and query processing for GAV and LAV.
- Given two data sources (schema, data guide, data, ...), integrate them using GAV or LAV.
- Define the notion of capabilities.
Page responsible: Patrick Lambrix
Last updated: 2021-08-21