J. McCusker et al. / What is a Knowledge Graph? 3
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9
10 10
11 11
12 12
13 13
14 14
15 15
16 16
17 17
18 18
19 19
20 20
21 21
22 22
23 23
24 24
25 25
26 26
27 27
28 28
29 29
30 30
31 31
32 32
33 33
34 34
35 35
36 36
37 37
38 38
39 39
40 40
41 41
42 42
43 43
44 44
45 45
46 46
47 47
48 48
49 49
50 50
51 51
knowledge management and graph databases. Our
purpose with this paper is to survey the evolv-
ing notion of a knowledge graph, to describe the
general space, and to provide an explicit opera-
tional description of a knowledge graph. We begin
with a review of recent definitions of knowledge
graphs, knowledge graph analysis and construc-
tion algorithms, and commercial, research, non-
profit, and government knowledge graphs. These
new knowledge graphs do not strictly adhere to
original knowledge graph theory [3], but instead
have followed a looser, more flexible definition.
We present a more descriptive view of current,
practical knowledge graphs, and discuss their po-
tential for evolution and impact.
2. Related Work
Rospocher, et al. present knowledge graphs as
collections of facts about entities, typically de-
rived from structured data sources such as Free-
base [4]. They cite a dearth of event representa-
tions in current knowledge graphs as a shortcom-
ing - limiting knowledge graphs to encyclopedic
items such as birth and death dates - primarily due
to the difficulty of obtaining temporal data about
entities in a structured manner. Recent surveys,
such as those by Hogenboom, et al. [5] and Deng,
et al. [6], provide overviews of numerous meth-
ods for event extraction from a variety of sources
including social media, news, academic publica-
tions, and even images and video, indicating that
there is a great interest in finding ways to interpret
and include such temporal data in a more struc-
tured format. Another review by Nickel et al. ex-
plores machine learning methods for knowledge
graphs, but limits their definition to directed la-
beled graphs, with the ability to optionally pre-
define the schema. They also review, but do not
take a position on, the use of the closed versus
open world assumptions.
van de Riet and Meersman [3], Stokman and de
Vries [7], and Zhang [8], present a formal theory
of knowledge graphs as a specialization of seman-
tic networks where meaning is expressed as struc-
ture, statements are unambiguous, and a limited
set of relation types are used. These requirements
also minimize redundancy within the knowledge
graph, which simplifies analytical operations (in-
cluding reasoning and queries). Popping explores
the use of knowledge graphs, and their challenges
at the time, in their use in network text analysis
[9]. Following Zhang, Popping defines the knowl-
edge graph as a type of semantic network that uses
only a few types of relations, but also asserts that
additional knowledge may be added to the graph.
Ehrlinger [10] selected some representative def-
initions that demonstrate the lack of a common
core understanding of the concept. Farber, et
al. [11] and Huang, et al. [12] define knowledge
graph as being an RDF graph. Paulheim [13]
argues that "knowledge graphs are supposed to
cover at least a major portion of the domains that
exist in the world, and are not supposed to be
restricted to only one domain." But while DB-
pedia or Wikidata are general knowledge graphs
and don’t focus on a single domain, this should
not mean that all knowledge graphs must be gen-
eral. On the contrary, we believe that knowledge
graphs created for specific domains such as Bi-
ology can be considered knowledge graphs if
they follow the other requirements. More recently,
many works report on automatically building
knowledge graphs out of textual medical knowl-
edge and medical records [14], [15], [16], [17].
3. A Definition of “Knowledge Graph”
One thing to note is that the knowledge graph
platforms that have been reviewed in this paper
do not strictly adhere to the definition of knowl-
edge graph that was set out in an de Riet and
Meersman [3], Stokman and de Vries [7], and
Zhang [8]. Since usage has evolved, it is appro-
priate to develop a definition that follows how
the term is currently used. Implicit in the name,
“knowledge graph,” is, of course, that a knowl-
edge graph represents knowledge, and does so us-
ing a graph structure. Stokman, de Vries [7], and
Zhang [8] posit useful definitions and require-
ments for knowledge graphs as a starting point: