Skip to main content
SpringerLink
Log in

A Method of Sharing Sentence Vectors for Opinion Triplet Extraction

  • Published:
Neural Processing Letters Aims and scope Submit manuscript

Abstract

The aspect-based sentiment analysis (ABSA) task mainly detects the sentiment polarities of aspect terms, or achieves the aspect-opinion co-extraction. Existing ABSA methods usually divide the task into two independent sub-tasks to execute separately, which will result in getting many meaningless pairs due to the following two reasons: the aspect-sentiment pairs extraction task has no reference opinion terms for the interaction, and the aspect-opinion co-extraction task has no reference to their corresponding sentiment dependencies. In recent years, the triplet extraction task is based on above problems with ABSA task to extract aspect terms, opinion terms and sentiment. However, existing triplet extraction methods still have insufficient interaction among the three sub-tasks and serious error propagation problems. Besides, actual sentences often have overlapping aspect words and opinion words. In this paper, we formulate the ABSA as an opinion triplet extraction (OTE) task under the multi-task learning framework. Meanwhile, a method of sharing sentence vectors for opinion triplet extraction (OTE-SSV) is proposed to enhance the extraction ability of text semantic representation and strengthen the interaction among the elements of triplet extraction. Unlike the existing OTE approaches that rely on a pipeline manner for multiple tasks, OTE-SSV uses a concurrent way to extract multiple sub-tasks, which greatly reduces the errors presenting in the previous subtask propagate into the next subtask extraction task. It is experimentally verified that OTE-SSV can also correctly and efficiently extract triplet in sentences with overlapping aspect words and opinion words. Experimental results on four ABSA semeval benchmarks show that F1 measures of OTE-SSV are 1% to 2% obviously superiorer to series of state-of-the-art technologies.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

Notes

  1. BIO tagging is a joint tagging strategy. In our method, an aspect or opinion label is set as B, I, O, where B, I, and O denote begin, inside, and outside of a span, respectively. For a sentence as:“The mushroom barley soup is amazing.”, it includes a named entity: mushroom barley soup. Instead of tagging the three words individually, we tag the label“food”to the entire phrase “mushroom barley soup”. This is a joint annotation. At last, the sentence is marked as: {O, B-POS, I-POS, I-POS, O, O, O} in terms of an aspect label, where the“the”=O,“mushroom”=B-POS , “barley”=I-POS,“soup”=I-POS,“is”=O,“amazing”=O,“.”=O. And {O, O, O, O, O, B, O} in terms of an opinion label, where the“the”=O, “mushroom”=O ,“barley”=O,“soup”=O,“is”=O,“amazing”=B,“.”=O. Here NEU, NEG, and POS are neutral, negative, and positive.

  2. A sentiment tag set as {NEU, NEG, POS} is unified into the aspect-sentiment label set {B-NEU, I-NEU, B-NEG, I-NEG, B-POS, I-POS, O}, where NEU, NEG, and POS are neutral, negative, and positive.

References

  1. Liu B (2012) Sentiment analysis and opinion mining. Synthesis lectures on human language technologies 5(1):1–167

    Article  Google Scholar 

  2. Chen Z, Qian T (2020) Relation-aware collaborative learning for unified aspect-based sentiment analysis. In: Proceedings of the 58th annual meeting of the association for computational linguistics. pp 3685–3694

  3. Wang W, Pan SJ, Dahlmeier D, Xiao X (2017) Coupled multi-layer attentions for co-extraction of aspect and opinion terms. In: Proceedings of the AAAI conference on artificial intelligence. Vol. 31

  4. Li X, Bing L, Li P, Lam W, Yang Z. (2018) Aspect term extraction with history attention and selective transformation. In: Proceedings of the 27th international joint conference on artificial intelligence. pp 4194–4200

  5. Zhang C, Li Q, Song D, Wang B (2020) A multi-task learning framework for opinion triplet extraction. In: Proceedings of the 2020 conference on empirical methods in natural language processing: Findings. pp 819–828

  6. Zhu X, Sobihani P, Guo H (2015) Long short-term memory over recursive structures. In: International conference on machine learning. pp 1604–1612. PMLR

  7. Haiyun Peng LX, Bing L, Huang F, Wei L, Si L (2020) Knowing what, how and why: a near complete solution for aspect-based sentiment analysis. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 34. No. 05. pp 8600–8607

  8. He R, Lee WS, Ng HT, Dahlmeier D (2019) An interactive multi-task learning network for end-to-end aspect-based sentiment analysis. In: Proceedings of the AAAI conference on artificial intelligence 33(01):6714–6721

  9. Luo H, Li T, Liu B, Zhang J (2019) DOER: dual cross-shared RNN for aspect term-polarity co-extraction. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. pp 591–601

  10. Lample G, Ballesteros M, Subramanian S, Kawakami K, Dyer C (2016) Neural architectures for named entity recognition. In: Proceedings of NAACL-HLT. pp 260–270

  11. Dozat T, Manning CD (2017) Deep biaffine attention for neural dependency parsing. In: 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings. pp 24–26

  12. Chaturvedi I, Ragusa E, Gastaldo P, Zunino R, Cambria E (2018) Bayesian network based extreme learning machine for subjectivity detection. J Franklin Inst 355(4):1780–1797

    Article  MathSciNet  MATH  Google Scholar 

  13. Manandhar S (2014)Semeval-2014 task 4: aspect based sentiment analysis. In: Proceedings of the 8th international workshop on semantic evaluation (SemEval 2014)

  14. Brun C, Popa DN, Roux C(2014) XRCE: hybrid classification for aspect-based sentiment analysis. In Nakov P, Zesch T, editors. In: Proceedings of the 8th International Workshop on Semantic Evaluation, SemEval@COLING 2014, Dublin, Ireland, August 23-24, 2014. pp 838–842. The Association for Computer Linguistics

  15. Tang H, Ji D, Li C, Zhou Q (2020) Dependency graph enhanced dual-transformer structure for aspect-based sentiment classification. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. pp 6578–6588

  16. Bin L, Liu Quan X, Jin ZQ, Peng Z (2017) Aspect-based sentiment analysis based on multi-attention CNN. J Comput Res Dev 54(8):1724

    Google Scholar 

  17. Xiao Y, Zhou G (2020) Syntactic edge-enhanced graph convolutional networks for aspect-level sentiment classification with interactive attention. IEEE Access 8:157068–157080

    Article  Google Scholar 

  18. Li X, Bing L, Li P, Lam W (2019) A unified model for opinion target extraction and target sentiment prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence 33(01):6714–6721

  19. Chen S, Liu J, Wang Y, Zhang W, Chi Z (2020) Synchronous double-channel recurrent network for aspect-opinion pair extraction. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, month = jul, address =Online, publisher =Association for Computational Linguistics. pp 6515–6524

  20. Zhao H, Huang L, Zhang R, Lu Q, Xue H (2020) A span-based multi-task learning framework for pair-wise aspect and opinion terms extraction. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, month = jul, address =Online, publisher =Association for Computational Linguistics. pp 3239–3248

  21. Huang L, Wang P, Li S, Liu T, Zhang X, Cheng Z, Yin D, Houfeng W (2021) First target and opinion then polarity: enhancing target-opinion correlation for aspect sentiment triplet extraction

  22. Zhang M, Zhang Y, Fu G (2017) End-to-end neural relation extraction with global optimization. In: Proceedings of the 2017 conference on empirical methods in natural language processing. pp 1730–1740

  23. Zheng S, Wang F, Bao H, Hao Y, Zhou P, Xu B (2017) Joint extraction of entities and relations based on a novel tagging scheme. In: ACL (1)

  24. Wadden D, Wennberg U, Luan Y, Hajishirzi H(2019) Entity, relation, and event extraction with contextualized span representations. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). pp 5788–5793

  25. Zhong Z, Chen D (2021) A frustratingly easy approach for joint entity and relation extraction. North American Association for Computational Linguistics (NAACL)

  26. Dai H, Song Y (2019) Neural aspect and opinion term extraction with mined rules as weak supervision. In: Proceedings of the 57th annual meeting of the association for computational linguistics. pp 5268–5277

  27. Hu M, Zhao S, Zhang L, Cai K, Su Z, Cheng R, Shen X (2019) CAN: constrained attention networks for multi-aspect sentiment analysis. In: Inui K, Jiang J, Ng V, Xiaojun Wan X (Eds) Proceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing, EMNLP-IJCNLP 2019, Hong Kong, China, November 3-7, 2019. pp 4600–4609. Association for Computational Linguistics

  28. Chen S, Liu J, Wang Y, Zhang W, Chi Z (2020) Synchronous double-channel recurrent network for aspect-opinion pair extraction. In: Proceedings of the 58th annual meeting of the association for computational linguistics. pp 6515–6524

  29. Zhao H, Huang L, Zhang R, Lu Q, et al (2020) SpanMlt: a span-based multi-task learning framework for pair-wise aspect and opinion terms extraction. In: Proceedings of the 58th annual meeting of the association for computational linguistics. pp 3239–3248

  30. Pennington J, Socher R, Manning CD (2014) Glove: global vectors for word representation. In: Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). pp 1532–1543

  31. Chung J, Gulcehre C, Cho K, Bengio Y (2014) Empirical evaluation of gated recurrent neural networks on sequence modeling. In: NIPS 2014 Workshop on Deep Learning, December 2014

  32. Bekoulis G, Deleu J, Demeester T, Develder C (2018) Joint entity recognition and relation extraction as a multi-head selection problem. Expert Syst Appl 114:34–45

    Article  Google Scholar 

  33. Pontiki M, Galanis D, Pavlopoulos J, Papageorgiou H, Androutsopoulos I, Manandhar S (2014) Semeval-2014 task 4: aspect based sentiment analysis. pages 27–35, Dublin, Ireland, August. Association for Computational Linguistics

  34. Pontiki M, Galanis D, Papageorgiou H, Manandhar S, Androutsopoulos I. (2015) Semeval-2015 task 12: aspect based sentiment analysis. In: Proceedings of the 9th international workshop on semantic evaluation (SemEval 2015). pp 486–495

  35. Pontiki M, Galanis D, Papageorgiou H, Androutsopoulos I, Manandhar S, Al-Smadi M, Al-Ayyoub M, Zhao Y, Qin B, De Clercq O, et al. (2016) Semeval-2016 task 5: aspect based sentiment analysis. In: International workshop on semantic evaluation. pp 19–30

  36. Wu Z, Ying C, Zhao F, Fan Z, Dai X, Xia R (2020) Grid tagging scheme for aspect-oriented fine-grained opinion extraction. Findings of the Association for Computational Linguistics: EMNLP. pp 2576–2585

  37. Wang Y, Hou Y, Che W, Liu T (2020) From static to dynamic word representations: a survey. Int J Mach Learn Cybern 11:4

    Article  Google Scholar 

  38. Mikolov T, Chen K, Corrado G, Dean J (2013) Efficient estimation of word representations in vector space, Computer Science

  39. Joulin A, Grave E, Mikolov T, Bojanowski P (2017) Bag of tricks for efficient text classification. In: Proceedings of the 15th conference of the European chapter of the association for computational linguistics, vol. 2, short papers. Association for Computational Linguistics, Valencia, Spain. p 427

  40. Ningsih AK, Hadiana AI (2021) Disaster tweets classification in disaster response using bidirectional encoder representations from transformer (BERT). booktitle=IOP Conference Series: Materials Science and Engineering, 1115, 1, 012032, organization=IOP Publishing

Download references

Acknowledgements

This work is supported by the National Key R &D Program of China (No. 2020AAA01051 01), the National Natural Science Foundation of China (Nos. 61976182, 61876157) and Sichuan Key R & D project (Nos. 2022YFH0020, 2021YFG0136, 2021YFG0312, 2021YFS0014).

Author information

Authors and Affiliations

  1. Southwest Jiaotong University, School of Computing and Artificial Intelligence, Chengdu, 611756, China

    Wenwen Liu, Jie Hu, Shengdong Du, Hongmei Chen & Fei Teng

Authors
  1. Wenwen Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jie Hu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Shengdong Du
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Hongmei Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Fei Teng
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jie Hu.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, W., Hu, J., Du, S. et al. A Method of Sharing Sentence Vectors for Opinion Triplet Extraction. Neural Process Lett 55, 751–772 (2023). https://doi.org/10.1007/s11063-022-10907-5

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11063-022-10907-5

Keywords

Search

Navigation