Knowledge graphs are being used for the detection of money laundering, insurance fraud, and other suspicious activities. Some recent work demonstrated how knowledge graphs are being used to study the impact of the COVID-19 outbreak on the economy. The fact that knowledge graphs are being used in more and more interdisciplinary problems calls for a reliable source of interdisciplinary knowledge. In this paper, we study the integration of knowledge graphs in the domains of economics, banking, and finance. Our integrated knowledge graph has over 610K nodes and 1.7 million edges. By performing statistical and graph-theoretical analysis, we demonstrate how the integration results in more entities with richer information. Its quality was examined by analyzing the subgraphs of the identity links and (pseudo-)transitive relations. Finally, we study the sources of error, and their refinement and discuss the benefit of our integrated graph.
1. Introduction
based on the dynamics of complex inter-connected
systems. Unfortunately, many sources of knowledge were
The 2008 financial crisis urged early detection of systemic developed independently of each other. Fusing these
inrisk to national and world economies in derivatives mar- dependent KGs could lead to a significantly richer source
kets. The relative size of these markets is a fundamental of knowledge which could improve the performance of
risk to geopolitical as well as economic security [
The WBG Taxonomy3 includes 3,882 concepts. It serves as a small classification schema which represents the concepts used to describe the World Bank Group’s topical
2https://fundontology.com/
3https://vocabulary.worldbank.org/PoolParty/wiki/taxonomy
knowledge domains and areas of expertise, providing ontology alignment and the set of correspondences is
an enterprise-wide, application-independent framework. called a mapping or an alignment.
In comparison, the Bank Regulation Ontology (BRO) is By integrating knowledge graphs of various domains,
much bigger and uses two industrial standards, namely we expect more entities and richer information for
entiFIBO and LKIF [
In this paper we study properties of integrated knowl- 1. the Financial Industry Business Ontology (we coledge graphs in the domain of economics, banking and lected the FIBO ontology using OWL and FIBO ifnance. Our results show that even though the integrated vocabulary using SKOS)5 knowledge graph has some errors which have been cre- 2. the Financial Regulation Ontology (FRO)6 ated due to minor mistakes, the overall usefulness has 3. the Hedge Fund Regulation (HFR) ontology7 been improved. Our contributions are: 4. the Legal Knowledge Interchange Format (LKIF) a) We integrate some knowledge graphs in the domain ontology8 of economics, banking, and finance and present the inte- 5. the Bank Regulation Ontology (BRO)9 grated knowledge graph consisting of over 610K entities 6. the Financial Instrument Global Identifier (FIGI) 10 and 1.7 million triples4. 7. the STW Thesaurus for Economics (and its mapb) We study how the integration can enrich the in- pings)11 formation of entities with some statistical and graph- 8. the Journal of Economic Literature (JEL) classifitheoretical analysis. cation system12
c) We discuss the source of error and its refinement of 9. the Fund Ontology13 the integrated knowledge graph for future use.
The paper is organised as follows: Section 2 presents Not all knowledge graphs are available: some are not
the knowledge graphs and their statistics. Section 3 open source (e.g., the Italian Ownership Graph [
The nodes can be IRIs, literals, or blank nodes. The 5The product version retrieved from https://edmconnect. edges are relations between nodes and their types in ifleedsmicnouTnucritll.eorgfo/fibrmoiantt)eraensdtgrhotutpps/fib:/o/e-dpmrocdouncntse/cfibto.e-domwclounc(1il4.o7rg/ the form of triples. Ontologies are semantic models of fibointerestgroup/fibo-products/fibo-voc (1 file in Turtle format) data that define the entities, their properties and types, respectively on 14th January, 2022. types and subtyping, as well as relations between entities. 632 Turtle files were retrieved from https://finregont.com/ An ontology can be represented as a knowledge graph. ontology-directory-files-prefixes/ on 14th Janurary, 2022.
An integrated knowledge graph G = ⟨V, E, L, l⟩ com/7o1n2tToluorgtlye-filfileess/woenre14rtehtrJiaenveudarfyro,2m02h2ttps://hedgefundontology. is a combination of a set of knowledge graphs 8Retrieved from http://www.estrellaproject.org/lkif-core/ {1, . . . , } where V = 1 ∪ . . . ∪ , E = 1 ∪ #download on 30th January, 2022. . . . ∪ , and L = 1 ∪ . . . ∪ . A function l : E → 2l 916 Turtle files were retrieved from https://bankontology.com/ assigns to each edge a set of labels, which is the union onto10lo4gRyD-dFirfileecstowryer-efilerse-ptrrieefixveesd/ ofrnom30thhttJpasn:u//awryw, w20.o2m2. g.org/spec/ of the labels: l() = 1() ∪ . . . ∪ (). For a given set FIGI/ on 22nd December, 2021. relations R, the subgraph is the graph GR with L = R. 11The paper used STW v9.12 based on the SKOS ontology. The When R = {}, GR = G. Often times, such an integra- ontology and its 9 mappings files were retrieved from https://zbw.eu/ tion requires the process of determining correspondences stw/version/latest/download/about.en.html on 30th Janurary, 2022. between concepts in ontologies. Such a process is called iden1t2iTfierhse/jTeul/ratbleofiluetwoans3re0ttrhieJvaendufarroym,2h0t2t1p.s://zbw.eu/beta/external_ 13The paper used 8 Turtle files retrieved from https:// fundontology.com/ontology-files/ on 28th December, 2021.
4The data and Python scripts are available at https://github.com/ 14https://agnos.ai/services shuaiwangvu/EcoFin-integrated. 15http://krrwebtools.cs.ox.ac.uk/logmap/ we used the version with mapping repair but not the aid of any reasoner. Unfortunately, FRO, BRO, and HFR failed to load due to parsing errors in some files they import.
Table 1 summarizes the number of pairs of entities generated by LogMap. Overall, 1,698 unique identity links of skos:exactMatch were added to the integrated graph.
All the knowledge graphs were first converted to
Turtle format and then used the RDFpro16 [
For the sake of speed, when studying properties of these knowledge graphs, we use files in their HDT format.
16http://rdfpro.fbk.eu/
entities can be enriched with some statistical analysis of graph structure (Section 3.1). We then examine identity links (e.g. skos:exactMatch) in the integrated graph G and their corresponding subgraphs (Section 3.2). Finally, we study transitive and pseudo-transitive relations such as concept generalisation (Section 3.3) followed by a discussion (Section 3.5).
We study how the information of entities can be enriched when combining diferent resources. When an entity is described in diferent domains, its in- and outdegree are expected to increase. Figure 1 illustrates the in-/out-degree of the knowledge graphs and the integrated knowledge graph. Both the in- and out-degrees of the integrated graph show a power-law distribution. Moreover, the figures show that the integration increases both the number of degrees in general and the number of nodes with high degrees, which demonstrates how this integration can enrich the information of entities. For example, lkif-core-norm:allowed_by has an outdegree of 7 in the integrated graph but the three graphs that contain information about it has out-degrees of 2, 5, and 1 respectively17.
A strongly connected component (SCC) of a directed graph is a maximal subgraph where there is a path between all pairs of vertices. A weakly connected component (WCC) is a subgraph of the original graph where all vertices are connected to each other by some path, ignoring the direction of edges. Table 3 summarizes the graph-theoretical statistics. Let maxSCC and maxWCC represent the number of nodes in the largest strongly connected component and weakly connected component respectively. In addition, we compute the fraction of nodes in the biggest SCC and WCC, denoted and respectively. The high values of in the table show that the graphs are mostly connected. More specifically, = 99.98% for the integrated graph, which is due to the overlapping domains of the knowledge graphs and the mappings. The low values of indicate that the underlying structure of these graphs is mostly hierarchical, especially that of JEL, BRO, and FIBO-vD.
considered identical and intended to refer to the same 17The prefix lkif-core-norm corresponds to the namespace http: //www.estrellaproject.org/lkif-core/norm.owl#. triples about skos:exactMatch. In addition, there are 8,172 triples about skos:relatedMatch, and 6,418 triples about skos:closeMatch. Figure 2 shows the frequency distribution of the weakly connected components in their corresponding subgraphs.
The largest two connected components of the
subgraph of owl:sameAs are with 8 and 6 entities
each. In contrast, the largest two connected components
of skos:exactMatch are much bigger, with 119 and
45 entities respectively. For skos:relatedMatch,
the largest weakly connected component consists of
21 entities. That of skos:closeMatch consists of
52 entities. A manual examination below shows that
there are errors in these large connected components.
The mis-use of these SKOS mapping properties can
have less implications than the owl:sameAs since
skos:exactMatch indicates only “a high degree of
confidence that the concepts can be used interchangeably
across a wide range of applications”[
on the definition of class subsumption, concept
generalisation, organisation composition, etc. Due to transitivity,
Figure 1: Distribution of in-/out-degree of nodes in knowledge entities in cycles imply some equivalence relation, which
graphs could be erroneous. Take lkif-core:component_of
for example. A triple specifies that “some thing is a
(functional) component of some other thing”. Entities in a
real-world entities. Typical identity links use relations cycle of lkif-core:component_of indicate that all
they are components of each other, which could be
errosuch as owl:sameAs and skos:exactMatch. We first
study identity links in G and their corresponding sub- neous. Some past work showed how strongly connected
graphs. In contrast to the statistics reported by Raad components can be used to locate errors when refining
et al., where owl:sameAs is much more popular than knowledge graphs [
There are in total 20 relations typed Our analysis also shows that the identity links come
owl:TransitiveProperty in G. We also study solely from two sources: the owl:sameAs triples are
the pseudo-transitive relations: those relations that from the FIBO-OWL knowledge graph, the triples
are not typed owl:TransitiveProperty but shows about skos:exactMatch, skos:closeMatch, and
transitivity in their intended semantics [
Take skos:broadMatch for example. A manual
analysis of the largest three SCC shows the edges could be 3.5. Discussion
erroneous. These SCCs are: a component with four
entities about plebiscite, referendum, and popular initiative; As shown above, this integration results in new
statistia component with three entities about insurance and pri- cal and graph-theoretical properties. Next, we compare
vate insurance; a component with three distinct entities how these problems exhibit in our graph and the
LODabout the CARICOM countries, Caribbean countries, and a-lot21 [
Let GB be the subgraph of the integrated graph G with a single ready-to-consume file. While our integrated
B = {skos:broader, skos:broaderMatch} and GN for knowledge graph has 1.7 million unique triples,
LOD-aN = {skos:narrower, skos:narrowerMatch}. Next, lot is much larger with 28.3 billion triples. For LOD-a-lot,
we combine the GB with the graph G’N, where G’N is a 356.9K edges out of 11.8 million edges of skos:broader
graph with each edge of G reversed in direction. After are involved in SCCs [
Our analysis shows that rdfs:subClassOf is a pop- of both owl:sameAs and skos:exactMatch are small ular relation with 47,597 triples. However, there is no and can be manually refined.
SCC with more than one component, which implies that the underlying class hierarchy is a directed acyclic graph. 4. Conclusion In addition, lkif-core:component, fro:divides19, and its inverse fro:divided_by are also popular tran- In this paper, we presented an integrated knowledge sitive relations. Finally, none of them has strongly con- graph in the domain of Economics, Finance, and Banking. nected components of size greater or equal to two. We demonstrated how the integrated graph has more entities with richer information. We discussed subgraphs of 3.4. Source of Error and Refinement (pseudo-)transitive and identity relations as well as their When tracing back to the sources of each edge, we found refinement. The overall usefulness has been improved despite minor errors introduced due to integration. ftrhoamtstkhorsee:bsroouarcdeesr: aSnTdWs,kJoEsL:,naanrdroFwIBeOr-avrDe. mWohstelny Our integrated knowledge graph can be used to evaluate data interoperability. Also, it can enrich the features combined with the subgraph of skos:broadMatch and of entities, which may increase the accuracy of pattern esnktoitsie:sn, atwrrooSwCMCastocfht,htrheeereenatriteieisn, atontdaltw44o SSCCCCss ooff ftowuor recognition using Machine Learning for the detection of entities. It is feasible that some domain experts manually takeovers, money laundering, insurance fraud, counterexamine all these small SCCs without employing any feiting, etc. Furthermore, it can also be used to improve refinement algorithm. the quality of suspicious activity reports, recommendation systems, conversational agents, etc.
19The prefix fro corresponds to the namespace http://finregont. com/fro/ref/LegalReference.ttl#. 20https://github.com/jrvosse/amalgame 21http://lod-a-lot.lod.labs.vu.nl/