DBpedia FlexiFusion the Best of Wikipedia &gt; Wikidata &gt; Your Data

Frey, Johannes; Hofer, Marvin; Obraczka, Daniel; Lehmann, Jens; Hellmann, Sebastian

doi:10.1007/978-3-030-30796-7_7

Cited by 14 publications

(13 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The distribution of property count among Wikidata entities follows a heavy-tailed distribution, with an average of 3.93 different properties per entity; This observation also applies to those entities that have corresponding Wikipedia pages across multiple language editions, with a slightly higher average of 4.96 property per entity [12]. The problem is compounded for entities that belong to rare classes.…”

Section: Introductionmentioning

confidence: 90%

Wiki2Prop: A Multimodal Approach for Predicting Wikidata Properties from Wikipedia

Luggen

Audiffren

Difallah

et al. 2021

Proceedings of the Web Conference 2021

View full text Add to dashboard Cite

Wikidata is rapidly emerging as a key resource for a multitude of online tasks such as Speech Recognition, Entity Linking, Question Answering, or Semantic Search. The value of Wikidata is directly linked to the rich information associated with each entity -that is, the properties describing each entity as well as the relationships to other entities. Despite the tremendous manual and automatic efforts the community invested in the Wikidata project, the growing number of entities (now more than 100 million) presents multiple challenges in terms of knowledge gaps in the graph that are hard to track. To help guide the community in filling the gaps in Wikidata, we propose to identify and rank the properties that an entity might be missing. In this work, we focus on entities with a dedicated Wikipedia page in any language to make predictions directly based on textual content. We show that this problem can be formulated as a multi-label classification problem where every property defined in Wikidata is a potential label. Our main contribution, Wiki2Prop, solves this problem using a multimodal Deep Learning method to predict which properties should be attached to a given entity, using its Wikipedia page embeddings. Moreover, Wiki2Prop is able to incorporate additional features in the form of multilingual embeddings and multimodal data such as images whenever available. We empirically evaluate our approach against the state of the art and show how Wiki2Prop significantly outperforms its competitors for the task of property prediction in Wikidata, and how the use of multilingual and multimodal data improves the results further. Finally, we make Wiki2Prop available as a property recommender system that can be activated and used directly in the context of a Wikidata entity page. CCS CONCEPTS• Human-centered computing → Wikis; • Information systems → Incomplete data; Network data models; • Computing methodologies → Supervised learning.This paper is published under the Creative Commons Attribution 4.0 International (CC-BY 4.0) license. Authors reserve their rights to disseminate the work on their personal and corporate Web sites with the appropriate attribution.

show abstract

Section: Introductionmentioning

confidence: 90%

Wiki2Prop: A Multimodal Approach for Predicting Wikidata Properties from Wikipedia

Luggen

Audiffren

Difallah

et al. 2021

Proceedings of the Web Conference 2021

View full text Add to dashboard Cite

show abstract

“…import, creation of documentation and reports, consistency tests and versioning. Although we were not able to find any announced public repository, the ontology organization structure is very similar to the one of DBpedia Databus [ 5 ] Archivo is based on.…”

Section: Related Workmentioning

confidence: 99%

“…DBpedia Archivo is built on top of the DBpedia Databus [ 5 ], which is inspired by Maven Central Repository. It uses the maven concepts publisher/group/artifact/version and ports them to a Linked Data platform, in order to manage data pipelines and enable automatic publishing and consumption of data.…”

Section: Archivo Platform Modelmentioning

confidence: 99%

“…In this paper, we present a web-scale ontology interface called DBpedia Archivo (acronym for ontology archive), that discovers, crawls and versions ontologies and archives as well as augments them on the DBpedia Databus [ 5 ]. The primary purpose of this interface is to help users/consumers to discover, access and validate/assess the quality/usability of ontologies in a unified way, while reducing the challenges and effort to spot and deal with the mentioned issues, such that they can focus on building stable and reliable applications.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DBpedia Archivo: A Web-Scale Interface for Ontology Archiving Under Consumer-Oriented Aspects

Frey

Streitmatter

Götz

et al. 2020

Semantic Systems. In the Era of Knowledge Graphs

Self Cite

View full text Add to dashboard Cite

While thousands of ontologies exist on the web, a unified system for handling online ontologies – in particular with respect to discovery, versioning, access, quality-control, mappings – has not yet surfaced and users of ontologies struggle with many challenges. In this paper, we present an online ontology interface and augmented archive called DBpedia Archivo, that discovers, crawls, versions and archives ontologies on the DBpedia Databus. Based on this versioned crawl, different features, quality measures and, if possible, fixes are deployed to handle and stabilize the changes in the found ontologies at web-scale. A comparison to existing approaches and ontology repositories is given .

show abstract

“…It can be seen that both DBpedia and Wikidata are the conversion of Wikipedia data into structured data using different methods [26]. However, some parts of Wikidata have been converted and incorporated into DBpedia Wikidata [27] and the ProFusion dataset [28].…”

Section: Manually Recording Facts Into the Knowledge Basementioning

confidence: 99%

TULIP: A Five-Star Table and List - From Machine-Readable to Machine-Understandable Systems

Nandakwang¹,

Chongstitvatana²

2020

Linked Open Data - Applications, Trends and Future Developments

View full text Add to dashboard Cite

Currently, Linked Data is increasing at a rapid rate as the growth of the Web. Aside from new information that has been created exclusively as Semantic Web-ready, part of them comes from the transformation of existing structural data to be in the form of five-star open data. However, there are still many legacy data in structured and semi-structured form, for example, tables and lists, which are the principal format for human-readable, waiting for transformation. In this chapter, we discuss attempts in the research area to transform table and list data to make them machine-readable in various formats. Furthermore, our research proposes a novel method for transforming tables and lists into RDF format while maintaining their essential configurations thoroughly. And, it is possible to recreate their original form back informatively. We introduce a system named TULIP which embodied this conversion method as a tool for the future development of the Semantic Web. Our method is more flexible compared to other works. The TULIP data model contains complete information of the source; hence it can be projected into different views. This tool can be used to create a tremendous amount of data for the machine to be used at a broader scale.

show abstract

DBpedia FlexiFusion the Best of Wikipedia > Wikidata > Your Data

Cited by 14 publications

References 9 publications

Wiki2Prop: A Multimodal Approach for Predicting Wikidata Properties from Wikipedia

Wiki2Prop: A Multimodal Approach for Predicting Wikidata Properties from Wikipedia

DBpedia Archivo: A Web-Scale Interface for Ontology Archiving Under Consumer-Oriented Aspects

TULIP: A Five-Star Table and List - From Machine-Readable to Machine-Understandable Systems

Contact Info

Product

Resources

About