Exploring Sequence-to-Sequence Learning in Aspect Term Extraction

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 3538–3547 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 3538 Exploring Sequence-to-Sequence Learning in Aspect Term Extraction Dehong Ma♣, Sujian Li♣, Fangzhao Wu♠, Xing Xie♠, Houfeng Wang♣ ♣MOE Key Lab of Computational Linguistics, Peking University, Beijing, 100871, China ♠ Microsoft Research Asia, Beijing, China {madehong, lisujian, wanghf}@pku.edu.cn wufangzhao@gmail.com, Xing.Xie@microsoft.com Abstract Aspect term extraction (ATE) aims at iden- tifying all aspect terms in a sentence and is usually modeled as a sequence labeling prob- lem. However, sequence labeling based meth- ods cannot make full use of the overall mean- ing of the whole sentence and have the lim- itation in processing dependencies between labels. To tackle these problems, we first explore to formalize ATE as a sequence-to- sequence (Seq2Seq) learning task where the source sequence and target sequence are com- posed of words and labels respectively. At the same time, to make Seq2Seq learning suit to ATE where labels correspond to words one by one, we design the gated unit networks to in- corporate corresponding word representation into the decoder, and position-aware attention to pay more attention to the adjacent words of a target word. The experimental results on two datasets show that Seq2Seq learning is effec- tive in ATE accompanied with our proposed gated unit networks and position-aware atten- tion mechanism. 1 Introduction Aspect term extraction (ATE) is a fundamental task in aspect-level sentiment analysis, and aims at extracting all aspect terms present in the sen- tences (Hu and Liu, 2004; Pontiki et al., 2014, 2015, 2016). For example, given a restaurant re- view “The staff is friendly, and their cheese pizza is delicious”, the ATE system should extract as- pect terms “staff” and “cheese pizza”. Early works focus on detecting the pre-defined aspects in a sentence (Hu and Liu, 2004; Zhuang et al., 2006; Popescu and Etzioni, 2007). Then, some works regard ATE as a sequence labeling task and utilize Hidden Markov Model (Jin et al., 2009) or Conditional Random Fields (Jin et al., 2009; Ma and Wan, 2010; Jakob and Gurevych, 2010; Liu et al., 2013) to extract all possible as- pect terms. With the development of deep learn- ing techniques, neural networks based methods (Wang et al., 2016; Liu et al., 2015; Li and Lam, 2017; Xu et al., 2018) have achieved good perfor- mances in ATE task, and they still treat ATE as a sequence labeling problem and extract more use- ful features surrounding a word. Obviously, the overall meaning of the sentence is important to predict the label sequence. For example, the word memory should be an aspect term in the laptop re- view “The memory is enough for use.”, but it is not an aspect term in the sentence “The memory is sad for me.”. However, sequence labeling meth- ods are not good at grasping the overall meaning of the whole sentence because they cannot read the whole sentence in advance. In addition, neural networks based …

Content truncated to 500 words