N³ - A Collection of Datasets for Named Entity Recognition and Disambiguation in the NLP Interchange Format

By Michael Röder, Ricardo Usbeck, Sebastian Hellmann, Daniel Gerber, Andreas Both. 2014

94 citations recorded by the source corpus.

Extracting Linked Data following the Semantic Web principle from unstructured sources has become a key challenge for scientific research. Named Entity Recognition and Disambiguation are two basic operations in this extraction process. One step towards the realization of the Semantic Web vision and the development of highly accurate tools is the availability of data for validating the quality of processes for Named Entity Recognition and Disambiguation as well as for algorithm tuning. This article presents three novel, manually curated and annotated corpora (N3). All of them are based on a free license and stored in the NLP Interchange Format to leverage the Linked Data character of our datasets.

Subfields: Artificial Intelligence, Information Systems, Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

Autoregressive Entity Retrieval (2020, methodology): “We evaluate on seven out-of-domain test sets: MSNBC, Derczynski (Der) (Der-czynski et al., 2015), KORE 50 (K50) (Hoffart et al., 2012), N3-Reuters-128 (R128), N3-RSS-500 (R500) (R ¨ oder et al., 2014), and OKE challenge 2015 and 2016 (OKE15 and OKE16) (Nuzzolese et al., 2015).”
CM3: A Causal Masked Multimodal Model of the Internet (2022, methodology): “For both setting we evaluate on seven test sets: MSNBC, Derczynski (Der) (Derczynski et al., 2015), KORE 50 (K50) (Hoffart et al., 2012), N3-Reuters-128 (R128), N3-RSS-500 (R500) (Röder et al., 2014), and OKE challenge 2015 and 2016 (OKE15 and OKE16) (Nuzzolese et al., 2015).”
GERBIL - Benchmarking Named Entity Recognition and Linking consistently (2017, methodology): “Moreover, we capitalize upon the uptake of publicly available, NIF based corpora over the last years [57, 52].”