Fine grained classification of named entities

The success of the Semantic Web depends on the availability of ontologies as well as on the proliferation of web pages annotated with metadata conforming to these ontologies. Thus, a crucial question is where to acquire these metadata. In this paper we propose PANKOW (Pattern-based Annotation through Knowledge on the Web), a method which employs an unsupervised, pattern-based approach to categorize instances with regard to an ontology. The approach is evaluated against the manual annotations of two human subjects. The approach is implemented in OntoMat, an annotation tool for the Semantic Web and shows very promising results.

show abstract

“…Other researches have considered this harder task such as Hahn and Schnattinger [15], Alfonseca and Manandhar [2] or Fleischman and Hovy [11].…”

Section: Related Workmentioning

confidence: 99%

“…Fleischmann and Hovy [11] address the classification of named entities into fine-grained categories. In particular, they categorize named entities denoting persons into the following 8 categories: athlete, politician/government, clergy, businessperson, entertainer/ artist, lawyer, doctor/scientist, police.…”

Section: Related Workmentioning

confidence: 99%

Towards the self-annotating web

Cimiano

Handschuh

Staab³

2004

Proceedings of the 13th International Conference on World Wide Web

242

170

View full text Add to dashboard Cite

show abstract

“…WordNet (Fellbaum, 1998), DBPedia (Auer et al, 2007), Freebase (Bollacker et al, 2008), BabelNet (Navigli and Ponzetto, 2012), among others) to automatically generate 'silver-standard' annotated corpora, without incurring the cost associated with gaining access to manually annotated corpora (e.g. Fleischman and Hovy (2002), Ling and Weld (2012), Nothman et al (2013)). We follow this general approach with the production of a large-scale automatically-created MWEntity resource extracted from BabelNet, used to distantly supervise our classifiers.…”

Section: Related Workmentioning

confidence: 99%

Multi-word Entity Classification in a Highly Multilingual Environment

Chesney¹,

Jacquet²,

Steinberger³

et al. 2017

Proceedings of the 13th Workshop on Multiword Expressions (MWE 2017)

View full text Add to dashboard Cite

This paper describes an approach for the classification of millions of existing multiword entities (MWEntities), such as organisation or event names, into thirteen category types, based only on the tokens they contain. In order to classify our very large in-house collection of multilingual MWEntities into an application-oriented set of entity categories, we trained and tested distantly-supervised classifiers in 43 languages based on MWEntities extracted from BabelNet. The best-performing classifier was the multi-class SVM using a TF.IDF-weighted data representation. Interestingly, one unique classifier trained on a mix of all languages consistently performed better than classifiers trained for individual languages, reaching an averaged F1-value of 88.8%. In this paper, we present the training and test data, including a human evaluation of its accuracy, describe the methods used to train the classifiers, and discuss the results.

show abstract

“…Washington the state, and Washington the city). The system is enhanced in Fleischman and Hovy (2002), for subtype classification of persons into 8 professions. Instead of using surface forms from the context, each context position is represented as the likelihood that it cooccurs with any of the 8 subtypes.…”

Section: Related Workmentioning

confidence: 99%

Fine-grained Dutch named entity recognition

Desmet

Hoste

2013

Lang Resources & Evaluation

View full text Add to dashboard Cite

This paper describes the creation of a fine-grained named entity annotation scheme and corpus for Dutch, and experiments on automatic main type and subtype named entity recognition. We give an overview of existing named entity annotation schemes, and motivate our own, which describes six main types (persons, organizations, locations, products, events and miscellaneous named entities) and finer-grained information on subtypes and metonymic usage. This was applied to a one-million-word subset of the Dutch SoNaR reference corpus. The classifier for main type named entities achieves a micro-averaged F-score of 84.91%, and is publicly available, along with the corpus and annotations.

show abstract

Fine grained classification of named entities

Cited by 130 publications

References 12 publications

Towards the self-annotating web

Towards the self-annotating web

Multi-word Entity Classification in a Highly Multilingual Environment

Fine-grained Dutch named entity recognition

Contact Info

Product

Resources

About