Named entity recognition with document-specific KB tag gazetteers

Abstract

We consider a novel setting for Named Entity Recognition (NER) where we have access to document-specific knowledge base tags. These tags consist of a canonical name from a knowledge base (KB) and entity type, but are not aligned to the text. We explore how to use KB tags to create document-specific gazetteers at inference time to improve NER. We find that this kind of supervision helps recognise organisations more than standard widecoverage gazetteers. Moreover, augmenting document-specific gazetteers with KB information lets users specify fewer tags for the same performance, reducing cost.

Neural Named Entity Recognition from Subword Units

“…NER is a widely studied problem, where methods have been characterized by the use of CRFs with heavy feature engineering, gazetteers and external knowledge resources (Finkel et al, 2005;Florian et al, 2003;Kazama and Torisawa, 2007;Klein et al, 2003;Lin and Wu, 2009;Radford et al, 2015;Ratinov and Roth, 2009;Zhang and Johnson, 2003). Ratinov and Roth (2009) (Bharadwaj et al, 2016;Chiu and Nichols, 2016;Collobert et al, 2011;Gillick et al, 2016;Huang et al, 2015;Lample et al, 2016;Ma and Hovy, 2016;dos Santos and Guimarães, 2015;Yadav et al, 2018;Yang et al, 2016).…”

A First Step Towards the Development of Yoruba Named Entity Recognition System

“…This paper has adopted the CRF model because it has proven itself as a successful algorithm in many sequence labelling tasks, including NER. It has been widely applied in notable previous research [14,15,7,16,17]. CRF is an undirected graphical model of the hidden markov model sort but which is conditionally trained.…”

A Study of the Importance of External Knowledge in the Named Entity Recognition Task

“…Following (Radford et al, 2015), after the first run of NED, we create a set of document-specific gazetteers derived from the disambiguated entities. This information helps in the second round to find new named entities that were previously missed.…”

“…To perform our study we use a linear chain CRF (Lafferty et al, 2001). CRFs are transparent and widely used for NER (Finkel et al, 2005;Jun'ichi and Torisawa, 2007;Ratinov and Roth, 2009;Passos et al, 2014;Radford et al, 2015;Luo et al, 2015). The entity-based component was implemented using the AIDA (Hoffart et al, 2011) entity disambiguation system.…”

“…NER is a widely studied problem. Most of previous work rely on the use of CRFs (Finkel et al, 2005;Jun'ichi and Torisawa, 2007;Ratinov and Roth, 2009;Passos et al, 2014;Radford et al, 2015;Luo et al, 2015). A recent trend has achieved particularly good results modeling NER as an end-to-end task using neural networks (dos Santos and Guimarães, 2015;Chiu and Nichols, 2016;Lample et al, 2016;Yang et al, 2016; Gillick et al, 2016).…”