Making Sense of Microposts (#Microposts2016) Named Entity rEcognition and Linking (NEEL) Challenge

Giuseppe Rizzo, Marieke van Erp, Julien Plu, Raphaël Troncy Making Sense of Microposts (#Microposts2016) Named Entity rEcognition and Linking (NEEL) Challenge.
In: Daniel Preoţiuc-Pietro, Danica Radovanović, Amparo E. Cano Basave, Katrin Weller, A.-S. Dadzie (eds.): Proceedings of the Workshop on Making Sense of Microposts (#Microposts2016),
Montréal,Canada, April 11th, 2016,
online: CEUR-WS.org/Vol-1691/microposts2016_neel-challenge-report.

This paper describes the (#Microposts2016) Named Entity rEcognition and Linking (NEEL) Challenge, held as a track of the Making Sense of Microposts Workshop co-located with the World Wide Web conference (WWW'2016). The challenge task comprised automatic linking and classification of entities appearing in different event streams of English tweets. Participants were invited to develop novel strategies for extracting entities in a tweet stream, typing them based on a set of pre-defined classes, and linking them to resources from the DBpedia 2015-04 knowledge base or NIL referents. The challenge attracted a lot of interest: 37 research teams expressed an intent to participate and signed the agreement to acquire the dataset. Six teams with different approaches took part in the final evaluation of the challenge task. The submissions covered joint linguistic and graph-based entity recognition and linking methods and sequential linguistic pipelines, where the two stages are separated, for addressing the challenge task. We describe the evaluation process and discuss the performance of the different approaches that have entered the 2016 NEEL Challenge. We also release, with this paper, the corpus consisting of manually annotated tweets.


Copyright © 2016 by the paper's authors |valid HTML5|