Abstract
The sentiment analysis task has been given great attention in the recent years, especially by enterprises and customers in commercial domain. In fact, companies tend to identify the customers’ opinions regarding their services, industrialized products, etc. In this context, the Aspect Based Sentiment Analysis (ABSA) was introduced to determine the clients’ viewpoints and extract the different aspects (e.g., price, quality, etc.) of entity (e.g., laptops) and assign them a sentiment polarity. In the present work, we are interested only in the aspect extraction (AE) task which is the most crucial and difficult task in the ABSA domain. We propose a hybrid method that combines the strengths of the linguistic knowledge and those of deep learning methods to solve the problem of AE for the French language. We also enhance this method by means of a new pruning algorithm which is mainly based on an out-domain dataset. The developed hybrid method has significantly improved the current state of the art and has given encouraging results when applied on respectively the Amazon mobile phone reviews (86.39% of F-measure) and the SemEval-2016 restaurant dataset (76.62% of F-measure).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The French lexical dictionary FEEL (French Expanded Emotion Lexicon) is composed of 14,128 opinion words: 8424 of them are positive and 5704 of them are negative.
- 2.
The value of support is chosen according to an empirical study that we effected on 2000 reviews. This study proves that the highly correlated aspect terms appear together for at least 20 times out of a total of 2000 transactions i.e. support 20/2000 = 0.01.
- 3.
The genism library was utilized to train our Word2Vec model on 100 epochs with a vector size of 100.
References
Ameen, B., Zohair, A., Ali, D., Rabeeh, A., Hussain, D.: Aspect based sentiment analysis for polarity estimation of customer reviews on Twitter. Comput. Mater. Continua 67(2), 2203–2225 (2020)
Baris, O., Akcayol, M.: A new topic modeling based approach for aspect extraction in aspect based sentiment analysis: SS-LDA. Expert Syst. Appl. 168, 114231 (2021)
Chauhan, G.S., Meena, Y.K., Gopalani, D., Nahta, R.: A two-step hybrid unsupervised model with attention mechanism for aspect extraction. Expert Syst. Appl., 161–113673 (2020)
Chen, Z., Mukherjee, A., Liu, B.: Aspect extraction with automated prior knowledge learning. In: Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, pp. 347–358 (2014)
Guang Q., Bing, L., Jiajun, B., Chun, C.: Expanding domain sentiment lexicon through double propagation. In: 31 International Joint Conference on Artificial Intelligence, vol. 9, pp. 1199–1204 (2009)
Manju, V., Gupta, D.: An unsupervised hierarchical rule-based model for aspect term extraction augmented with pruning strategies. Procedia Comput. Sci. 171, 22–31 (2020)
Minqing, H., Liu, B.: Mining and summarizing customer reviews. In: Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 168–177 (2004)
Nihel, P., Erwan, K.: Analyse de sentiments à base d’aspects par combinaison de réseaux profonds: application à des avis en Français . In: conférence sur le Traitement Automatique des Langues, pp. 303–310 (2018)
Qian, L., Bing, L., Yuanlin, Z., Doo, K., Zhiqiang, G.: Improving opinion aspect extraction using semantic similarity and aspect associations. In: Thirtieth AAAI Conference on Artificial Intelligence (2016)
Rakesh, A., Srikant, R.: Fast algorithms for mining association rules. In: Proceedings of 20th International Conference Very Large Data Bases, VLDB, pp. 478–499 (1994)
Samaneh, M., Ester, M.: Opinion digger: an unsupervised opinion miner from unstructured product reviews. In: Proceedings of the 19th ACM International Conference on Information and Knowledge Management, pp. 1825–1828. (2020)
Sebastian, R., Ghaffari, P., Breslin, J.: Insight-1 at semeval-2016 task 5: Deep learning for multilingual aspect-based sentiment analysis. arXiv preprint arXiv:1609.02748 (2016)
Zhang, C., Li, Q., Song, D., Wang, B.: A multi-task learning framework for opinion triplet extraction. arXiv preprint arXiv:2010.01512 (2020)
Liang, Y., Meng, F., Zhang, J., Chen, Y., Xu, J., Zhou, J.: A dependency syntactic knowledge augmented interactive architecture for end-to-end aspect-based sentiment analysis. Neurocomputing 454, 291–302 (2021)
Wu, C., Wu, F., Wu, S., Yuan, Z., Huang, Y.: A hybrid unsupervised method for aspect term and opinion target extraction. Knowl.-Based Syst. 148, 66–73 (2018)
Bhamare, B.R., Prabhu, J.: A supervised scheme for aspect extraction in sentiment analysis using the hybrid feature set of word dependency relations and lemmas. PeerJ Comput. Sci. 7, e347 (2021)
Ruidan, H., Wee, S.L.: An interactive multi-task learning network for end-to-end aspect-based sentiment analysis. arXiv preprint arXiv:1906.06906 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Hammi, S., Hammami, S.M., Belguith, L.H. (2022). Aspect Term Extraction Improvement Based on a Hybrid Method. In: Ceci, M., Flesca, S., Masciari, E., Manco, G., Raś, Z.W. (eds) Foundations of Intelligent Systems. ISMIS 2022. Lecture Notes in Computer Science(), vol 13515. Springer, Cham. https://doi.org/10.1007/978-3-031-16564-1_9
Download citation
DOI: https://doi.org/10.1007/978-3-031-16564-1_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16563-4
Online ISBN: 978-3-031-16564-1
eBook Packages: Computer ScienceComputer Science (R0)