TY - JOUR
AU - 
AB - Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification † † † † § Minghao Hu , Yuxing Peng , Zhen Huang , Dongsheng Li , Yiwei Lv National University of Defense Technology, Changsha, China University of Macau, Macau, China {huminghao09,pengyuxing,huangzhen,dsli}@nudt.edu.cn Abstract Typically, the whole task can be decoupled into two subtasks. Since opinion targets are not given, Open-domain targeted sentiment analysis aims we need to first detect the targets from the in- to detect opinion targets along with their senti- put text. This subtask, which is usually denoted ment polarities from a sentence. Prior work as target extraction, can be solved by sequence typically formulates this task as a sequence tagging methods (Jakob and Gurevych, 2010; Liu tagging problem. However, such formulation et al., 2015; Wang et al., 2016a; Poria et al., 2016; suffers from problems such as huge search space and sentiment inconsistency. To ad- Shu et al., 2017; He et al., 2017; Xu et al., 2018). dress these problems, we propose a span-based Next, polarity classification aims to predict the extract-then-classify framework, where multi- sentiment polarities over the extracted target en- ple opinion targets are directly extracted from tities (Jiang et al., 2011; Dong et al., 
TI - Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification
JF - Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
DO - 10.18653/v1/p19-1051
DA - 2019-01-01
UR - https://www.deepdyve.com/lp/unpaywall/open-domain-targeted-sentiment-analysis-via-span-based-extraction-and-kAhROxg5nG
DP - DeepDyve
ER -