Knowledge Graphs Systems Activities

Exercises

Exercise 1.1. Identify which of the following satisfies the definition of a knowledge  graph introduced in this chapter. 

(a) A data graph defined among data items representing real-world entities. 

(b) A schema graph defined among classes in a schema.

(c) A process graph representing the steps of a process, their order, and branching conditions. 

(d) A parse tree is an ordered, rooted tree that represents the syntactic structure of a string according to some context-free grammar.

(e) An entity relationship diagram shows the relationships of entity sets stored in a database.

Exercise 1.2. Which of the following counts as a well-defined meaning of the labels in a knowledge graph?

(a) Names of the labels in a human understandable language.

(b) Everything in (a) plus a documentation string that explains the label in sufficient detail.

(c) Embeddings calculated for the relation names over a large corpus of text.

(d) Everything in (a) plus a specification in a formal language.

(e) Everything in (b) plus a specification in a formal language.

Exercise 1.3. Identify which of the following statements about knowledge graphs are true.

(a) Knowledge graphs are the only way to achieve data integration in enterprises.

(b) Edges in a knowledge graph are like the links between web documents except that the edges have semantically defined labels.

(c) A knowledge graph is the best representation for recording the output of NLP and vision algorithms.

(d) Semantic networks were the earliest knowledge graphs in AI.

(e) Understanding is to brain as a knowledge graph is to AI.