Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-1528
Title: Transfer learning and sentence level features for named entity recognition on tweets
Authors : von Däniken, Pius
Cieliebak, Mark
Proceedings: Proceedings of the 3rd Workshop on Noisy User-generated Text
Volume(Issue) : 3
Pages : 166
Pages to: 171
Conference details: 3rd Workshop on Noisy User-generated Text (W-NUT), Copenhagen, September 7th, 2017
Publisher / Ed. Institution : Association for Computational Linguistics
Issue Date: 7-Sep-2017
License (according to publishing contract) : Licence according to publishing contract
Type of review: Peer review (Publication)
Language : English
Subjects : Named Entity Recogintion; NER
Subject (DDC) : 004: Computer science
005: Computer programming, programs and data
Abstract: We present our system for the WNUT 2017 Named Entity Recognition challenge on Twitter data. We describe two modi- fications of a basic neural network architecture for sequence tagging. First, we show how we exploit additional labeled data, where the Named Entity tags differ from the target task. Then, we propose a way to incorporate sentence level features. Our system uses both methods and ranked second for entity level annotations, achieving an F1-score of 40.78, and second for surface form annotations, achieving an F1- score of 39.33.
Departement: School of Engineering
Organisational Unit: Institute of Applied Information Technology (InIT)
Publication type: Conference Paper
DOI : 10.21256/zhaw-1528
URI: https://digitalcollection.zhaw.ch/handle/11475/1854
Appears in Collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
W17-4422.pdf232.63 kBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.