Towards a Definition of Knowledge Graphs - Leitura de Artigo

Ehrlinger, Lisa and Wolfram Wöß. “Towards a Definition of Knowledge Graphs.” SEMANTiCS (2016).

ABSTRACT

The prerequisite for widespread academic and commercial adoption of a concept or technology is a common understanding, based ideally on a definition that is free from ambiguity. We tackle this issue by discussing and defining the term knowledge graph, considering its history and diversity in interpretations and use.

 

1. INTRODUCTION

 

Other definitions may lead to the assumption that knowledge graph is a synonym for any graph-based knowledge representation (cf. [12, 16]). We argue that such a definition is not enough for an adequate application of knowledge graphs, since it does not enforce a minimum set of requirements a KG has to fulfill. Thus, even a simple graph-based vocabulary could be published as knowledge graph.  

 

2. SELECTED DEFINITIONS

 
Knowledge graphs have been in the focus of research since 2012 resulting in a wide variety of published descriptions and definitions. 

 

“A knowledge graph (i) mainly describes real world entities and their interrelations, organized in a graph, (ii) defines possible classes and relations of entities in a schema, (iii) allows for potentially interrelating arbitrary entities with each other and (iv) covers various topical domains.”
Paulheim [16] ... listed in his survey of knowledge graph refinement the minimum set of characteristics that must be present to distinguish knowledge graphs from other knowledge collections, which basically restricts the term to any graph-based knowledge representation.

 

** A estrutura em grafo, a forma como os dados são representados, seria o diferenciados entre KG e KB **

 
“Knowledge graphs are large networks of entities, their semantic types, properties, and relationships between entities.”
Journal of Web Semantics [12]

 
“Knowledge graphs could be envisaged as a network of all kind things which are relevant to a specific domain or to an organization. They are not limited to abstract concepts and relations but can also contain instances of things like documents and datasets.”
Semantic Web Company [3]

 

Vague descriptions of knowledge graphs were published in the announcement of a special issue on knowledge graphs by the Journal of Web Semantics and by the Semantic Web Company. Both definitions could equally well describe an ontology or – even more generally – any kind of semantic knowledge representation and do not even enforce a graph structure. In addition, size is highlighted as an essential characteristic, which is reflected by phrases such as “large networks” or “vast networks”
[11], while it remains unclear what “large” means in this context.

 

** Na Era Big Data, o quanto de volume é necessário para ser considerado Big ainda é nebuloso ** 

** Não deixam claro ser em grafo mas quando falam em ser em rede estão deixando isso implícito **

** Ontologias também tem instâncias então não seria isso que diferente a Ontologia de um KG ** 

 
“We define a Knowledge Graph as an RDF graph. An RDF graph consists of a set of RDF triples where each RDF triple (s, p, o) is an ordered set of the following RDF terms: a subject s ∈ U ∪ B, a predicate p ∈ U , and an object o ∈ U ∪ B ∪ L. An RDF term is either a URI u ∈ U , a blank node b ∈ B, or a literal l ∈ L.”
Färber et al. [7] ... defined a knowledge graph as an Resource Description Framework (RDF) graph and stated that the term KG was coined by Google to describe any graph-based knowledge base (KB) [7]. Although this definition is the only formal one, it contradicts with more general definitions as it explicitly requires the RDF data model.

 

** Restrito ao modelo RDF **

 
“[...] systems exist, [...], which use a variety of techniques to extract new knowledge, in the form of facts, from the web. These facts are interrelated, and hence, recently this extracted knowledge has been referred to as a knowledge graph.”
Pujara et al. [17] ... did not provide a concise definition, but rather described the characteristics of knowledge graphs. Unlike the other definitions, which focus solely on the inner structure of the KG, they highlighted the importance of an automatic extraction system.

 

** A capacidade de inferir novo conhecimento, revelar conhecimento implícito, seria uma característica dos KGs, mas ontologias também podem inferir conhecimento usando uma máquina de regras **

 

3. KNOWLEDGE GRAPH APPLICATIONS

 

In the 1980s, researchers from the University of Groningen and the University of Twente in the Netherlands initially introduced the term knowledge graph to formally describe their knowledge-based system that integrates knowledge from different sources for representing natural language [10, 15]. The authors proposed KGs with a limited set of relations and focus on qualitative modeling including human interaction, which clearly contrasts with the idea of KGs that has been widely discussed in recent years.

 

** Primeiras ocorrências do KG que não consideram as mesmas características esperadas de um KG **

 

In 2012, Google introduced the Knowledge Graph as a semantic enhancement of Google’s search function that does not match strings, but enables searching for “things”, in other words, real-world objects [18]. .... Since 2012, the term knowledge graph is also used to describe a family of applications. Frequently mentioned implementations are DBPedia, YAGO (Yet Another Great Ontology), Freebase, Wikidata, Yahoo’s semantic search assistant tool Spark, Google’s Knowledge Vault, Microsoft’s Satori and Facebook’s entity graph [7, 14, 16, 11].  

 

** Estado da Arte em termos de aplicações que usam KGs mas seriam mesmo KGs? **

 

4. TERMINOLOGICAL ANALYSIS AND DEFINITION

 

the terms knowledge graph and knowledge base are <Have Been> used interchangeably ... leads to the misleading assumption that the term knowledge graph is a synonym for knowledge base, which is itself often used as synonym for ontology.

 

** KG = KB ~= Ontologia ** 

** Muitas aplicações que usam/usaram ontologia só exploraram o lado terminológico como hierarquia de classes/conceitos e dentro do mesmo domínio, não exploraram inferências com instâncias em "grande" volume **

 

According to Akerkar and Sajja [1], a knowledge-based system uses artificial intelligence to solve problems, and it consists of two parts: a knowledge base and an inference engine. The knowledge base is a dataset with formal semantics that can contain different kinds of knowledge, for example, rules, facts, axioms, definitions, statements, and primitives [4].

 

** KB não tem só dados, tem regras, definições, restrições, etc ... que permitem inferir mais conhecimento **

 

An ontology is as a formal, explicit specification of a shared conceptualization that is characterized by high semantic expressiveness required for increased complexity [9]. Ontological representations allow semantic modeling of knowledge, and are therefore commonly used as knowledge bases in artificial intelligence (AI) applications, for example, in the context of knowledge-based systems. Application of an ontology as knowledge base facilitates validation of semantic relationships and derivation of  conclusions from known facts for inference (i.e., reasoning) [9]. We explicitly emphasize that an ontology does not differ from a knowledge base, although ontologies are sometimes erroneously classified as being at the same level as database schemas [6]. In fact, an ontology consists not only of classes and properties (e.g., owl:ObjectProperty and owl:DatatypeProperty), but can also hold instances (i.e., the population of the ontology).

 

** Ontologia também pode ter instâncias e regras, apesar de pouco explorado **

 

Thus, the difference between a knowledge graph and an ontology could be interpreted either as a matter of quantity (e.g., a large ontology), or of extended requirements (e.g., a built-in reasoner that allows new knowledge to be derived). 

Focusing on existing automatically generated “knowledge graphs”, we can identify a further essential characteristic: collection, extraction, and integration of information from external sources extends a pure knowledge-based system with the concept of integration systems.

 

 

 

** Dois requisitos: volume (?), obtido através da integração de diversas fontes, e um reasoner embutido **

 

A knowledge graph acquires and integrates information into an ontology and applies a reasoner to derive new knowledge. 

 

This definition aligns with the assumption that a knowledge graph is somehow superior and more complex than a knowledge base (e.g., an ontology) because it applies a reasoning engine to generate new knowledge and integrates one or more information sources. Consequently, a manually created knowledge graph that does not support integration aspects is a plain knowledge base or knowledge-based system if it provides reasoning capabilities.

 

** Wikidata é criada manualmente mas também recebe carga de dados **

 

The definition does not take the quantity aspect (size) into account, especially with respect to a large ABox of the ontology, since it is not clear what can be considered “large”. 

 

Furthermore, the question arises what constitutes the difference between the Semantic Web and knowledge graphs. Smaller KGs, for example, enterprise knowledge graphs, can be clearly differentiated from the Semantic Web because of their restricted domain. ... the Semantic Web could be interpreted as the most comprehensive knowledge graph, or – conversely – a knowledge graph that crawls the entire web could be interpreted as self-contained Semantic Web.