Noise Robust Named Entity Understanding for Voice Assistants
Computation and Language
2021-08-11 v3 Artificial Intelligence
Machine Learning
Abstract
Named Entity Recognition (NER) and Entity Linking (EL) play an essential role in voice assistant interaction, but are challenging due to the special difficulties associated with spoken user queries. In this paper, we propose a novel architecture that jointly solves the NER and EL tasks by combining them in a joint reranking module. We show that our proposed framework improves NER accuracy by up to 3.13% and EL accuracy by up to 3.6% in F1 score. The features used also lead to better accuracies in other natural language understanding tasks, such as domain classification and semantic parsing.
Cite
@article{arxiv.2005.14408,
title = {Noise Robust Named Entity Understanding for Voice Assistants},
author = {Deepak Muralidharan and Joel Ruben Antony Moniz and Sida Gao and Xiao Yang and Justine Kao and Stephen Pulman and Atish Kothari and Ray Shen and Yinying Pan and Vivek Kaul and Mubarak Seyed Ibrahim and Gang Xiang and Nan Dun and Yidan Zhou and Andy O and Yuan Zhang and Pooja Chitkara and Xuan Wang and Alkesh Patel and Kushal Tayal and Roger Zheng and Peter Grasch and Jason D. Williams and Lin Li},
journal= {arXiv preprint arXiv:2005.14408},
year = {2021}
}
Comments
NAACL 2021 Industry Track