An Insight Towards Prompt And Learning Based Sentiment Analysis And Classification With Feature Representation
DOI:
https://doi.org/10.29284/68pf7307Keywords:
sentiment analysis, prompt-based model, aspect model, information, training model.Abstract
Sentiment analysis generally aims to ascertain a text's general sentiment polarity, which is a general strategy that doesn't particularly target any one target. On the other hand, aspect-based sentiment analysis has attracted a lot of interest lately. It is a more thorough approach that determines sentiment by looking at particular elements in the text. To accomplish this job, the majority of methods have historically relied on deep learning in conjunction with attention processes. DL-based models, however, can be intricate and need a lot of training. Learning-based devices have gained popularity recently due to their enhanced performance. Even if these models are successful, they are unable to distinguish between the multiple logical relationships among the data's characteristics and fail to accurately model them. This work suggests a hybrid Markov Long Short Term Memory (M-LSTM) method known as prompt-enhanced Sentiment Analysis in response to this. This technique employs the mask to represent mutual information between aspects and finds training data that is most comparable to the input text. Moreover, when conducting sentiment analysis on several aspects during inference, the suggested model only requires one forward pass. Additionally, the language representation model is utilized to improve the suggested M-LSTM method's performance. Extensive experiments were carried out to verify this strategy. Here, prompt-enhanced Sentiment Analysis, a novel and effective model, is presented. In this model, prompts from datasets that contain reciprocal information between aspects are merged. Cloze-style fine-tuning is used in prompt learning which functions similarly to pretraining. Consequently, it makes better use of pretraining data and greatly improves few-shot performance. To further enhance performance, the language representation model is also applied.
References
[1] Alamanda MS. Aspect-based sentiment analysis search engine for social media data. CSI Trans ICT. 2020;8(2):193–7.
[2] Alqaryouti O, Siyam N, Monem AA, Shaalan K. Aspect-based sentiment analysis using smart government review data. Applied Computing and Informatics. 2019. p 1–12.
[3] Aslam N, Rustam F, Lee E, Washington PB, Ashraf I. Sentiment analysis and emotion detection on cryptocurrency related Tweets using ensemble LSTM-GRU Model. IEEE Access. 2022;10:39313–24.
[4] Ayyub K, Iqbal S, Munir EU, Nisar MW, Abbasi M. Exploring diverse features for sentiment quantifcation using machine learning algorithms. IEEE Access. 2020;8:142819–31.
[5] Bie Y, Yang Y. A multitask multiview neural network for end-to-end aspect-based sentiment analysis. Big Data Mining Analytics. 2021;4(3):195–207.
[6] Chandra R, Kulkarni V. Semantic and sentiment analysis of selected Bhagavad Gita translations using BERT-based language framework. IEEE Access. 2022;10:21291–315.
[7] Datta S, Chakrabarti S. Aspect based sentiment analysis for demonetization tweets by optimized recurrent neural network using fre fy-oriented multi-verse optimizer. Sadhana. 2021;46(2):1–23
[8] Flyand. A hybrid framework for sentiment analysis using genetic algorithm based feature reduction. IEEE Access. 2019;7:14637–52.
[9] Hao Y, Mu T, Hong R, Wang M, Liu X, Goulermas JY. Cross-domain sentiment encoding through stochastic word embedding. IEEE Trans Knowl Data Eng. 2020;32(10):1909–22
[10] Hitkul S, Shahid S, Singhal D, Mahata P, Kumarguru RR, Shah. Aspect-Based Sentiment Analysis of Financial Headlines and Microblogs. In: Deep Learning-Based Approaches for Sentiment Analysis, Springer; 2020. pp 111–137
[11] Huang F, Li X, Yuan C, Zhang S, Zhang J, Qiao S. Attention-emotion-enhanced convolutional LSTM for sentiment analysis. IEEE Trans Neural Netw Learn Syst. 2022;33(9):4332–45.
[12] Khan A, Gul MA, Zareei M, Biswal RR, Zeb A, Naeem M, Saeed Y, Salim N. Movie review summarization using supervised learning and graph-based ranking algorithm. Comput Intell Neurosci. 2020;2020:1–14.
[13] Khan J, Jeong BS. Summarizing customer review based on product feature and opinion. In: 2016 International Conference on Machine Learning and Cybernetics (ICMLC), 2016. p. 158–65.
[14] Kumar R, Pannu HS, Malhi AK. Aspect-based sentiment analysis using deep networks and stochastic optimization. Neural Comput Appl. 2020;32(8):3221–35.
[15] Li N, Chow CY, Zhang JD. SEML: a semi-supervised multi-task learning framework for aspect-based sentiment analysis. IEEE Access. 2020;8:189287–97.
[16] Londhe A, Rao PVRDP. Aspect Based Sentiment Analysis – An Incremental Model Learning Approach Using LSTM-RNN. In: Singh, M., Tyagi, V., Gupta, P.K., Flusser, J., Ören, T., Sonawane, V.R. (eds) Advances in Computing and Data Sciences. ICACDS 2021. Communications in Computer and Information Science, vol 1440. 2021; Springer, Cham.
[17] Lu Q, Zhu Z, Zhang G, Kang S, Liu P. Aspect-gated graph convolutional networks for aspect-based sentiment analysis. Appl Intell. 2021;51(7):4408–19.
[18] Moussa ME, Mohamed EH, Haggag MH. A survey on opinion summarization techniques for social media. Fut Comput Inform J. 2018;3(1):82–109.
[19] Muhammad PF, Kusumaningrum R, Wibowo A. Sentiment Analysis Using Word2vec and Long Short-Term Memory (LSTM) for Indonesian Hotel Reviews. Proc Int Conf Computer Sci Computat Intell Procedia Computer Sci. 2021;179:728–35.
[20] Munuswamy S, Saranya MS, Ganapathy S, Muthurajkumar S, Kannan A. Sentiment analysis techniques for social mediabased recommendation systems. Nat Acad Sci Lett. 2021;44(3):281–7.
[21] Oyebode O, Alqahtani F, Orji R. Using machine learning and thematic analysis methods to evaluate mental health apps based on user reviews. IEEE Access. 2020;8:111141–58.
[22] Prathi JK, Raparthi PK, Gopalachari MV. Real-time aspect-based sentiment analysis on consumer reviews. Data Eng Commun Technol. 2020;1079:801–10.
[23] Saif H, Fernández M, He Y, Alani H. Evaluation datasets for twitter sentiment analysis: a survey and a new dataset, the STSGold. ESSEM@AI*IA. 2013.
[24] Schouten K, Frasincar F. Survey on aspect-level sentiment analysis. IEEE Trans Knowl Data Eng. 2016;28(3):813–30.
[25] Schouten K, van der Weijde O, Frasincar F, Dekker R. Supervised and unsupervised aspect category detection for sentiment analysis with co-occurrence data. IEEE Trans Cybern. 2018;48(4):1263–75
[26] Shahana PH, Omman B. Evaluation of features on sentimental analysis. Procedia Computer Sci. 2015;46:1585–92.
[27] Shams M, Khoshavi N, Baraani-Dastjerdi A. LISA: language-independent method for aspect-based sentiment analysis. IEEE Access. 2020;8:31034–44.
[28] Singh M, Jakhar AK, Pandey S. Sentiment analysis on the impact of coronavirus in social life using the BERT model. Soc Netw Anal Min. 2021;11(1):11–33. H
[29] Wang D. Coarse alignment of topic and sentiment: a unifed model for cross-lingual sentiment classifcation. IEEE Trans Neural Netw Learn Syst. 2021;32:736–47.
[30] Wang L, Niu J, Yu S. SentiDif: combining textual information and sentiment difusion patterns for twitter sentiment analysis. IEEE Trans Knowl Data Eng. 2020;32(10):2026–39
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
