Skip to main content
Springer Nature Link
Log in

Aspect based sentiment analysis with instruction tuning and external knowledge enhanced dependency graph

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Aspect-Based Sentiment Analysis (ABSA) is generally defined as a fine-grained task in Natural Language Processing (NLP). Recently, the integration of the Large Language Model (LLM) and Graph Convolutional Network (GCN) has been widely studied to excavate the underlying contextual information and support the sentiment polarity prediction. However, in existing research, the LLM is usually employed directly to generate the contextual feature representation without any specific instructions, which is not suitable for learning the domain language corpus. In addition, the existing works usually fuse the contextual feature and graph feature by GCN simply, and it ignores further specific processing to highlight the sentiment representations before the model’s final outputting. To tackle these two imperfections, this work proposes a novel ABSA model Instruction Tuning-based Graph Convolutional Network (ITGCN) to implement the subtask of predicting sentiment polarities\(^\textrm{R2}\), which leverages the instructed LLM to generate the task-oriented contextual representation and the GCN to exploit the external affective knowledge-assisted syntactic features. In the proposed ITGCN, firstly, the inputting sentence is reconstructed with the designed task-specific instructions, which tell the LLM what is the target in the input. Secondly, this work’s dependency graph, before being processed by GCN, is weighted by the affective knowledge extracted from SenticNet. This kind of dependency graph is endowed with affective information, which is closer to the intention of the related study. Finally, to learn more structured knowledge, a bi-layer sentiment representation module is proposed and utilized to enhance the feature representation. To validate the effectiveness of the proposed ITGCN, extensive experiments have been conducted on five public and available datasets. The proposed ITGCN achieves competitive performance and outperforms the selected state-of-the-art baselines, obviously.

Graphical abstract

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

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

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability and access

The datasets analysed during the current study are available from the corresponding author on reasonable request (https://github.com/zhangzheng1997/SSEGCN-ABSA).

Notes

  1. In this work, the spaCy toolkit is used to derive the dependency tree of the review: https://spacy.io/.

References

  1. Alkatheiri MS (2022) Artificial intelligence assisted improved human-computer interactions for computer systems. Comput Electr Eng 101(107):950

    Google Scholar 

  2. An W, Tian F, Chen P et al (2023) Aspect-based sentiment analysis with heterogeneous graph neural network. IEEE Trans Comput Soc Syst 10(1):403–412. https://doi.org/10.1109/TCSS.2022.3148866

    Article  Google Scholar 

  3. Cao Y, Tang Y, Du H et al (2023) Heterogeneous reinforcement learning network for aspect-based sentiment classification with external knowledge. IEEE Trans Affect Comput pp 1–14. https://doi.org/10.1109/TAFFC.2022.3233020

  4. Cheng LC, Chen YL, Liao YY (2022) Aspect-based sentiment analysis with component focusing multi-head co-attention networks. Neurocomputing 489:9–17

    Article  Google Scholar 

  5. Cui L, Wu Y, Liu J et al (2021) Template-based named entity recognition using bart. In: Findings of the association for computational linguistics: ACL-IJCNLP 2021 pp 1835–1845

  6. Dai J, Yan H, Sun T et al (2021) Does syntax matter? a strong baseline for aspect-based sentiment analysis with Roberta. In: Proceedings of the 2021 conference of the North American chapter of the association for computational linguistics: human language technologies, pp 1816–1829

  7. Deng J, Ren F (2021) Hierarchical network with label embedding for contextual emotion recognition. Research 2021. https://doi.org/10.34133/2021/3067943. https://spj.science.org/doi/abs/10.34133/2021/3067943. https://arxiv.org/abs/https://spj.science.org/doi/pdf/10.34133/2021/3067943

  8. Devlin J, Chang MW, Lee K et al (2019) Bert: pre-training of deep bidirectional transformers for language understanding. In: Proceedings of the 2019 Conference of the North American chapter of the association for computational linguistics: human language technologies, vol 1 (Long and Short Papers), pp 4171–4186

  9. Feng S, Wang B, Yang Z et al (2022) Aspect-based sentiment analysis with attention-assisted graph and variational sentence representation. Knowl-Based Syst 258(109):975

    Google Scholar 

  10. García-Díaz P, Sanchez-Berriel I, Pontiel-Martín D et al (2023) A novel flexible feature extraction algorithm for Spanish tweet sentiment analysis based on the context of words. Expert Syst Appl 212:118817. https://api.semanticscholar.org/CorpusID:249303298

  11. Gu T, Zhao H, Li M (2022) Effective inter-aspect words modeling for aspect-based sentiment analysis. Appl Intell 53:4366–4379. https://api.semanticscholar.org/CorpusID:249532759

  12. Gu T, Zhao H, He Z et al (2023) Integrating external knowledge into aspect-based sentiment analysis using graph neural network. Knowl-Based Syst 259(110):025

    Google Scholar 

  13. Gu T, Zhao H, Li M (2023) Effective inter-aspect words modeling for aspect-based sentiment analysis. Appl Intell 53(4):4366–4379

    Article  Google Scholar 

  14. Han X, Zhao W, Ding N et al (2022) Ptr: prompt tuning with rules for text classification. AI Open 3:182–192

    Article  Google Scholar 

  15. Hannan MA, How DNT, Lipu MSH et al (2021) Soc estimation of li-ion batteries with learning rate-optimized deep fully convolutional network. IEEE Trans Power Electron 36:7349–7353. https://api.semanticscholar.org/CorpusID:229646875

  16. Hou X, Huang J, Wang G et al (2021) Selective attention based graph convolutional networks for aspect-level sentiment classification. In: Proceedings of the fifteenth workshop on graph-based methods for natural language processing (TextGraphs-15), pp 83–93

  17. Hu M, Peng Y, Huang Z et al (2019) Open-domain targeted sentiment analysis via span-based extraction and classification. In: Proceedings of the 57th annual meeting of the association for computational linguistics, pp 537–546

  18. Li XL, Liang P (2021) Prefix-tuning: optimizing continuous prompts for generation. In: Proceedings of the 59th annual meeting of the association for computational linguistics and the 11th International joint conference on natural language processing (vol 1: Long Papers), pp 4582–4597

  19. Liang B, Su H, Gui L et al (2022) Aspect-based sentiment analysis via affective knowledge enhanced graph convolutional networks. Knowl-Based Syst 235(107):643

    Google Scholar 

  20. Liu Y, Ott M, Goyal N et al (2019) Roberta: a robustly optimized bert pretraining approach. arXiv:1907.11692 Retrieved from https://arxivorg/abs/190711692

  21. Lu Q, Zhu Z, Zhang G et al (2021) Aspect-gated graph convolutional networks for aspect-based sentiment analysis. Appl Intell 51:4408 – 4419. https://api.semanticscholar.org/CorpusID:234181768

  22. Ma Y, Peng H, Cambria E (2018) Targeted aspect-based sentiment analysis via embedding commonsense knowledge into an attentive lstm. In: Proceedings of the AAAI conference on artificial intelligence

  23. Ma Y, Peng H, Khan T et al (2018) Sentic lstm: a hybrid network for targeted aspect-based sentiment analysis. Comput Cogn 10:639–650. https://api.semanticscholar.org/CorpusID:3876403

  24. Ma Y, Song R, Gu X et al (2022) Multiple graph convolutional networks for aspect-based sentiment analysis. Appl Intell 53:12985 – 12998. https://api.semanticscholar.org/CorpusID:252751388

  25. Mencarini E, Rapp A, Tirabeni L et al (2019) Designing wearable systems for sports: a review of trends and opportunities in human-computer interaction. IEEE Trans Human-Mach Syst 49(4):314–325

    Article  Google Scholar 

  26. Phan MH, Ogunbona PO (2020) Modelling context and syntactical features for aspect-based sentiment analysis. In: Proceedings of the 58th annual meeting of the association for computational linguistics, pp 3211–3220

  27. Radford A, Wu J, Child R et al (2019) Language models are unsupervised multitask learners. OpenAI blog 1(8):9

    Google Scholar 

  28. Rani S, Kumar P (2021) Aspect-based sentiment analysis using dependency parsing. Transactions on Asian and Low-Resource Language Information Processing 21:1 – 19. https://api.semanticscholar.org/CorpusID:245153835

  29. Song Y, Wang J, Jiang T, et al (2019) Attentional encoder network for targeted sentiment classification. International Conference on Artificial Neural Networks abs/1902.09314. https://api.semanticscholar.org/CorpusID:67855317

  30. Sun C, Huang L, Qiu X (2019) Utilizing bert for aspect-based sentiment analysis via constructing auxiliary sentence. In: Proceedings of NAACL-HLT, pp 380–385

  31. Vaswani A, Shazeer N, Parmar N et al (2017) Attention is all you need. Adv Neural Inf Process Syst 30

  32. Wang K, Shen W, Yang Y et al (2020) Relational graph attention network for aspect-based sentiment analysis. In: Proceedings of the 58th annual meeting of the association for computational linguistics, pp 3229–3238

  33. Wang X, Li F, Zhang Z et al (2021) A unified position-aware convolutional neural network for aspect based sentiment analysis. Neurocomputing 450:91–103

    Article  Google Scholar 

  34. Wang Y, Yang N, Miao D et al (2022) Dual-channel and multi-granularity gated graph attention network for aspect-based sentiment analysis. Appl Intell pp 1–13

  35. Wu H, Shi X (2022) Adversarial soft prompt tuning for cross-domain sentiment analysis. In: Proceedings of the 60th annual meeting of the association for computational linguistics (vol 1: Long Papers), pp 2438–2447

  36. Wu H, Zhang Z, Shi S, et al (2021) Phrase dependency relational graph attention network for aspect-based sentiment analysis. Knowl Based Syst 236:107736. https://api.semanticscholar.org/CorpusID:244482737

  37. Wu H, Huang C, Deng S (2023) Improving aspect-based sentiment analysis with knowledge-aware dependency graph network. Inf Fusion 92:289–299

    Article  Google Scholar 

  38. Xu L, Pang X, Wu J et al (2023) Learn from structural scope: improving aspect-level sentiment analysis with hybrid graph convolutional networks. Neurocomputing 518:373–383

    Article  Google Scholar 

  39. Zhang B, Li X, Xu X et al (2020) Knowledge guided capsule attention network for aspect-based sentiment analysis. IIEEE/ACM Trans Audio Speech Lang Process 28:2538–2551. https://api.semanticscholar.org/CorpusID:221590623

  40. Zhang B, Li X, Xu X et al (2020) Knowledge guided capsule attention network for aspect-based sentiment analysis. IEEE/ACM Trans Audio Speech Lang Process 28:2538–2551

    Article  Google Scholar 

  41. Zhang C, Li Q, Song D (2019) Aspect-based sentiment classification with aspect-specific graph convolutional networks. 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 4568–4578

  42. Zhao A, Yu Y (2021) Knowledge-enabled bert for aspect-based sentiment analysis. Knowl-Based Syst 227(107):220

    Google Scholar 

  43. Zhao G, Luo Y, Chen Q, et al (2023) Aspect-based sentiment analysis via multitask learning for online reviews. Knowl-Based Syst 110326

  44. Zhao M, Yang J, Shang F (2023) Dependency-enhanced graph convolutional networks for aspect-based sentiment analysis. Neural Comput & Applic 35:14195 –14211. https://api.semanticscholar.org/CorpusID:257748316

  45. Zhao Z, Tang M, Tang W et al (2022) Graph convolutional network with multiple weight mechanisms for aspect-based sentiment analysis. Neurocomputing 500:124–134

    Article  Google Scholar 

  46. Zhao Z, Tang M, Zhao FR et al (2022) Incorporating semantics, syntax and knowledge for aspect based sentiment analysis. Appl Intell 53:16138 – 16150. https://api.semanticscholar.org/CorpusID:254200878

  47. Zhong Q, Ding L, Liu J et al (2023) Knowledge graph augmented network towards multiview representation learning for aspect-based sentiment analysis. IEEE Trans Knowl Data Eng pp 1–14. https://doi.org/10.1109/TKDE.2023.3250499

  48. Zhou J, Chen Q, Huang JX et al (2020) Position-aware hierarchical transfer model for aspect-level sentiment classification. Inf Sci 513:1–16

  49. Zhou J, Huang J, Hu Q et al (2020) Is position important? deep multi-task learning for aspect-based sentiment analysis. Appl Intell 50:3367–3378. https://api.semanticscholar.org/CorpusID:219331102

  50. Zhou Y, Kang X, Ren F (2023) Prompt consistency for multi-label textual emotion detection. IEEE Trans Affect Comput pp 1–10. https://doi.org/10.1109/TAFFC.2023.3254883

  51. Zhu Z, Zhang D, Li L et al (2023) Knowledge-guided multi-granularity gcn for absa. Inf Process Manag 60(2):103223

    Article  Google Scholar 

Download references

Acknowledgements

This research was funded by the National Natural Science Foundation of China under Grant 62176084 and Grant 62176083, and in part by the Fundamental Research Funds for the Central Universities of China under Grant PA2022GDSK0066 and Grant PA2022GDSK0068.

Author information

Authors and Affiliations

  1. School of Computer and Information, Hefei University of Technology, Danxia Road, Hefei, 230601, Anhui, China

    Xuefeng Shi, Min Hu & Piao Shi

  2. Anhui Province Key Laboratory of Affective, Hefei University of Technology, Danxia Road, Hefei, 230601, Anhui, China

    Xuefeng Shi, Min Hu & Piao Shi

  3. School of Computer Science and Engineering, University of Electronic Science and Technology of China, Xiyuan Avenue, Chengdu, 611731, Sichuan, China

    Fuji Ren

  4. Graduate School of Information Science & Technology, The University of Tokyo, Kitamura, Tokyo, 113-8654, Japan

    Satoshi Nakagawa

Authors
  1. Xuefeng Shi
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Min Hu
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Fuji Ren
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Piao Shi
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Satoshi Nakagawa
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Xuefeng Shi and Fuji Ren prepared the whole plan and conducted the related experiments. Xuefeng Shi, Piao Shi and Min Hu wrote the main manuscript text, and Xuefeng Shi and Satoshi Nakagawa prepared figures, and Xuefeng Shi and Piao Shi prepared tables. All authors reviewed the manuscript.

Corresponding author

Correspondence to Xuefeng Shi.

Ethics declarations

Competing interests

Not Applicable

Ethical and informed consent for data used

Not Applicable

Additional information

Publisher's Note

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

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Shi, X., Hu, M., Ren, F. et al. Aspect based sentiment analysis with instruction tuning and external knowledge enhanced dependency graph. Appl Intell 54, 6415–6432 (2024). https://doi.org/10.1007/s10489-024-05492-0

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-024-05492-0

Keywords