Character-level Based Conference Named Entity Recognition Using Bi-LSTM

DOI

10.2991/cnci-19.2019.11How to use a DOI?

Keywords

Conference named entity recognition, character, BiLSTM, viterbi, open domain.

Abstract

For name entity recognition in specific field, a character-level bidirectional LSTM(Bidirectional Long Short-Term Memory, BiLSTM-based conference named entity recognition method is proposed. Introducing word embedding as the input of BiLSTM. each character in a sentence is mapped from a one-hot vector to a low-dimensional dense character embedding through word embedding. After the BiLSTM layer, viterbi algorithm is used to decode the output of BiLSTM. The results show that the precision rate, the recall rate and F value can reach 89.17%, 90.24% and 89.70%, indicating the effectiveness and practicability of the method of conference named entity recognition for open domain data.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Rui Xiong
AU  - Chensheng Wu
AU  - Li Xiong
AU  - Luyu Li
PY  - 2019/05
DA  - 2019/05
TI  - Character-level Based Conference Named Entity Recognition Using Bi-LSTM
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 83
EP  - 88
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.11
DO  - 10.2991/cnci-19.2019.11
ID  - Xiong2019/05
ER  -