A deep learning-based bilingual Hindi and Punjabi named entity recognition system using enhanced word embeddings

作者：

Highlights：

• Development of enhanced word embeddings for bilingual NER system is a novel attempt.

• Proposed work is the first attempt to develop a bilingual Hindi-Punjabi NER system.

• Our study reveals effectiveness of different embeddings for Hindi and Punjabi text.

• Blending of Bi-GRU and CNN model using EWE improves the performance of NER system.

• We determine that the proposed approach can be used for any resource-scarce language.

摘要

•Development of enhanced word embeddings for bilingual NER system is a novel attempt.•Proposed work is the first attempt to develop a bilingual Hindi-Punjabi NER system.•Our study reveals effectiveness of different embeddings for Hindi and Punjabi text.•Blending of Bi-GRU and CNN model using EWE improves the performance of NER system.•We determine that the proposed approach can be used for any resource-scarce language.

论文关键词：Named entity recognition (NER),Word embeddings,FastText,Bidirectional Gated Recurrent Unit (Bi-GRU),Natural Language Processing (NLP),Convolutional Neural Networks (CNN)

论文评审过程：Received 10 November 2020, Revised 9 February 2021, Accepted 9 October 2021, Available online 19 October 2021, Version of Record 29 October 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107601