REL: An Entity Linker Standing on the Shoulders of Giants

Hulst, Johannes M. van; Hasibi, Faegheh; Dercksen, Koen; Balog, Krisztian; Vries, Arjen P. de

doi:10.1145/3397271.3401416

Cited by 71 publications

(70 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) A variant was used where the candidate generation is enriched with historical spellings of Wikidata entities. (3) The last variant used an existing tool [115], which included contextual similarity and co-occurrence probabilities of mentions and Wikipedia articles. In the tool, the final disambiguation is based on the ment-norm method by Le and Titov [66].…”

Section: Language Model-based Approachesmentioning

confidence: 99%

Survey on English Entity Linking on Wikidata: Datasets and approaches

Cedric

Lehmann

Usbeck

2022

View full text Add to dashboard Cite

Wikidata is a frequently updated, community-driven, and multilingual knowledge graph. Hence, Wikidata is an attractive basis for Entity Linking, which is evident by the recent increase in published papers. This survey focuses on four subjects: (1) Which Wikidata Entity Linking datasets exist, how widely used are they and how are they constructed? (2) Do the characteristics of Wikidata matter for the design of Entity Linking datasets and if so, how? (3) How do current Entity Linking approaches exploit the specific characteristics of Wikidata? (4) Which Wikidata characteristics are unexploited by existing Entity Linking approaches? This survey reveals that current Wikidata-specific Entity Linking datasets do not differ in their annotation scheme from schemes for other knowledge graphs like DBpedia. Thus, the potential for multilingual and time-dependent datasets, naturally suited for Wikidata, is not lifted. Furthermore, we show that most Entity Linking approaches use Wikidata in the same way as any other knowledge graph missing the chance to leverage Wikidata-specific characteristics to increase quality. Almost all approaches employ specific properties like labels and sometimes descriptions but ignore characteristics such as the hyper-relational structure. Hence, there is still room for improvement, for example, by including hyper-relational graph embeddings or type information. Many approaches also include information from Wikipedia, which is easily combinable with Wikidata and provides valuable textual information, which Wikidata lacks.

show abstract

Section: Language Model-based Approachesmentioning

confidence: 99%

Survey on English Entity Linking on Wikidata: Datasets and approaches

Cedric

Lehmann

Usbeck

2022

View full text Add to dashboard Cite

show abstract

“…Preprocessing step. We identify NE mentions in podcast titles and descriptions and link them to Wikipedia entities using the Radboud Entity Linker (REL) system [10]. The REL system is based on multiple modules in pipeline specific to different sub-tasks: 1) the detection of NE mentions using Flair [1], a SOTA Named Entity Recognition (NER) framework using contextualized word embeddings; 2) the disambiguation of the identified entity against a list of possible Wikipedia candidates and its linking to the final candidate.…”

Section: Ne-informed Corpus Embedding (Neice)mentioning

confidence: 99%

Topic Modeling on Podcast Short-Text Metadata

Valero¹,

Baranes²,

Epure³

2022

Preprint

View full text Add to dashboard Cite

Podcasts have emerged as a massively consumed online content, notably due to wider accessibility of production means and scaled distribution through large streaming platforms. Categorization systems and information access technologies typically use topics as the primary way to organize or navigate podcast collections. However, annotating podcasts with topics is still quite problematic because the assigned editorial genres are broad, heterogeneous or misleading, or because of data challenges (e.g. short metadata text, noisy transcripts). Here, we assess the feasibility to discover relevant topics from podcast metadata, titles and descriptions, using topic modeling techniques for short text. We also propose a new strategy to leverage named entities (NEs), often present in podcast metadata, in a Non-negative Matrix Factorization (NMF) topic modeling framework. Our experiments on two existing datasets from Spotify and iTunes and Deezer, a new dataset from an online service providing a catalog of podcasts, show that our proposed document representation, NEiCE, leads to improved topic coherence over the baselines. We release the code for experimental reproducibility of the results 1 .

show abstract

“…We include four state-of-the-art entity linkers developed for documents and queries: REL [21], Blink [23], Genre [4], and ELQ [23]. We run these annotators with high-recall score thresholds, preserving score information for downstream applications, which is important for entity-based retrieval models [8].…”

Section: Query Entity Annotationmentioning

confidence: 99%

How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset

Mackie

Dalton

Yates

2021

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

Deep Learning Hard (DL-HARD) is a new annotated dataset designed to more effectively evaluate neural ranking models on complex topics. It builds on TREC Deep Learning (DL) topics by extensively annotating them with question intent categories, answer types, wikified entities, topic categories, and result type metadata from a commercial web search engine. Based on this data, we introduce a framework for identifying challenging queries. DL-HARD contains fifty topics from the official DL 2019/2020 evaluation benchmark, half of which are newly and independently assessed. We perform experiments using the official submitted runs to DL on DL-HARD and find substantial differences in metrics and the ranking of participating systems. Overall, DL-HARD is a new resource that promotes research on neural ranking methods by focusing on challenging and complex topics.

show abstract

REL: An Entity Linker Standing on the Shoulders of Giants

Cited by 71 publications

References 21 publications

Survey on English Entity Linking on Wikidata: Datasets and approaches

Survey on English Entity Linking on Wikidata: Datasets and approaches

Topic Modeling on Podcast Short-Text Metadata

How Deep is your Learning: the DL-HARD Annotated Deep Learning Dataset

Contact Info

Product

Resources

About