Transformation Networks for Target-Oriented Sentiment Classification
Refereed conference paper presented and published in conference proceedings

Full Text

Times Cited

Other information
AbstractTarget-oriented sentiment classification aims at classifying sentiment polarities over individual opinion targets in a sentence. RNN with attention seems a good fit for the characteristics of this task, and indeed it achieves the state-of-the-art performance. After re-examining the drawbacks of attention mechanism and the obstacles that block CNN to perform well in this classification task, we propose a new model to overcome these issues. Instead of attention, our model employs a CNN layer to extract salient features from the transformed word representations originated from a bi-directional RNN layer. Between the two layers, we propose a component to generate target-specific representations of words in the sentence, meanwhile incorporate a mechanism for preserving the original contextual information from the RNN layer. Experiments show that our model achieves a new state-of-the-art performance on a few benchmarks.
Acceptance Date15/07/2018
All Author(s) ListXin Li, Lidong Bing, Wai Lam, Bei Shi
Name of ConferenceThe 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018)
Start Date of Conference15/07/2018
End Date of Conference20/07/2018
Place of ConferenceMelbourne, Australia
Country/Region of ConferenceAustralia
Proceedings TitleProceedings of the 56th Annual Meeting of the Association for Computational Linguistics
Pages946 - 956
LanguagesEnglish-United Kingdom

Last updated on 2021-23-02 at 01:28