In the recent years, different Web knowledge graphs, both free and commercial, have been created. While Google coined the term "Knowledge Graph" in 2012, there are also a few openly available knowledge graphs, with DBpedia, YAGO, and Freebase being among the most prominent ones. Those graphs are often constructed from semi-structured knowledge, such as Wikipedia, or harvested from the web with a combination of statistical and linguistic methods. The result are large-scale knowledge graphs that try to make a good trade-off between completeness and correctness. In order to further increase the utility of such knowledge graphs, various refinement methods have been proposed, which try to infer and add missing knowledge to the graph, or identify erroneous pieces of information. In this article, we provide a survey of such knowledge graph refinement approaches, with a dual look at both the methods being proposed as well as the evaluation methodologies used.

Knowledge graph refinement: A survey of approaches and evaluation methods

In this paper we provide a comprehensive introduction to knowledge graphs,
which have recently garnered significant attention from both industry and
academia in scenarios that require exploiting diverse, dynamic, large-scale
collections of data. After some opening remarks, we motivate and contrast
various graph-based data models and query languages that are used for knowledge
graphs. We discuss the roles of schema, identity, and context in knowledge
graphs. We explain how knowledge can be represented and extracted using a
combination of deductive and inductive techniques. We summarise methods for the
creation, enrichment, quality assessment, refinement, and publication of
knowledge graphs. We provide an overview of prominent open knowledge graphs and
enterprise knowledge graphs, their applications, and how they use the
aforementioned techniques. We conclude with high-level future research
directions for knowledge graphs.

Knowledge Graphs

Type information is very valuable in knowledge bases. However, most large open knowledge bases are incomplete with respect to type information, and, at the same time, contain noisy and incorrect data. That makes classic type inference by reasoning difficult. In this paper, we propose the heuristic link-based type inference mechanism SDType, which can handle noisy and incorrect data. Instead of leveraging T-box information from the schema, SDType takes the actual use of a schema into account and thus is also robust to misused schema elements.

/pdf/type-inference-on-noisy-rdf-data-3gg52lvg91.pdf

Type Inference on Noisy RDF Data

Linked Data on the Web is either created from structured data sources (such as relational databases), from semi-structured sources (such as Wikipedia), or from unstructured sources (such as text). In the latter two cases, the generated Linked Data will likely be noisy and incomplete. In this paper, we present two algorithms that exploit statistical distributions of properties and types for enhancing the quality of incomplete and noisy Linked Data sets: SDType adds missing type statements, and SDValidate identifies faulty statements. Neither of the algorithms uses external knowledge, i.e., they operate only on the data itself. We evaluate the algorithms on the DBpedia and NELL knowledge bases, showing that they are both accurate as well as scalable. Both algorithms have been used for building the DBpedia 3.9 release: With SDType, 3.4 million missing type statements have been added, while using SDValidate, 13,000 erroneous RDF statements have been removed from the knowledge base.

/pdf/improving-the-quality-of-linked-data-using-statistical-2z9lyfofho.pdf

Improving the Quality of Linked Data Using Statistical Distributions

A machine reader is a tool able to transform natural language text to formal structured knowledge so as the latter can be interpreted by machines, according to a shared semantics. FRED is a machine reader for the semantic web: its output is a RDF/OWL graph, whose design is based on frame semantics. Nevertheless, FRED’s graph are domain and task independent making the tool suitable to be used as a semantic middleware for domainor task-specific applications. To serve this purpose, it is available both as REST service and as Python library. This paper provides details about FRED’s capabilities, design issues, implementation and evaluation.

/pdf/semantic-web-machine-reading-with-fred-14rsbzwe26.pdf

Semantic Web Machine Reading with FRED

We have implemented a novel approach for robust ontology design from natural language texts by combining Discourse Representation Theory (DRT), linguistic frame semantics, and ontology design patterns We show that DRT-based frame detection is feasible by conducting a comparative evaluation of our approach and existing tools Furthermore, we define a mapping between DRT and RDF/OWL for the production of quality linked data and ontologies, and present FRED, an online tool for converting text into internally well-connected and linked-data-ready ontologies in web-service-acceptable time

Knowledge extraction based on discourse representation theory and linguistic frames

We present Tipalo, an algorithm and tool for automatically typing DBpedia entities. Tipalo identifies the most appropriate types for an entity by interpreting its natural language definition, which is extracted from its corresponding Wikipedia page abstract. Types are identified by means of a set of heuristics based on graph patterns, disambiguated to WordNet, and aligned to two top-level ontologies: WordNet supersenses and a subset of DOLCE+DnS Ultra Lite classes. The algorithm has been tuned against a golden standard that has been built online by a group of selected users, and further evaluated in a user study.

/pdf/automatic-typing-of-dbpedia-entities-4eq60gbp8w.pdf

Automatic typing of DBpedia entities

FRED is an online tool for converting text into internally well-connected and quality linked-data-ready ontologies in web-service-acceptable time. It implements a novel approach for ontology design from natural language sentences. In this paper we present a demonstration of such tool combining Discourse Representation Theory (DRT), linguistic frame semantics, and Ontology Design Patterns (ODP). The tool is based on Boxer which implements a DRT-compliant deep parser. The logical output of Boxer enriched with semantic data from Verbnet or Framenet frames is transformed into RDF/OWL by means of a mapping model and a set of heuristics following ODP best-practice [5] of OWL ontologies and RDF data design.

/pdf/fred-from-natural-language-text-to-rdf-and-owl-in-one-click-gs3oomb3lf.pdf

FRED: From Natural Language Text to RDF and OWL in One Click

Aemoo is a Web application supporting exploratory search over the Semantic Web. Through a simple keyword-based search interface, users can query Aemoo for information about any entity, which is then collected by aggregating knowledge from diverse sources such as linked data, Wikipedia, Twitter, and Google News. Such aggregation is performed according to cognitively-sound principles through the exploitation of knowledge patterns, and by exploiting semantic relations as well as interpreting hypertext links. Aemoo provides users with an effective summary of knowledge about an entity, including explanations that clarify its relevance, and presents it through a user-friendly interface that supports exploration of further knowledge.

Francesco Draicchio

Papers

Semantic Web Machine Reading with FRED

Knowledge extraction based on discourse representation theory and linguistic frames

Automatic typing of DBpedia entities

FRED: From Natural Language Text to RDF and OWL in One Click

Aemoo: Exploratory Search based on KnowledgePatterns over the Semantic Web