Transformer-Based Semantic Self-Attention Regression for the Evaluation of Customer Satisfaction in Social Media Data
Received: 18 June 2025 | Revised: 23 August 2025 and 9 September 2025 | Accepted: 11 September 2025 | Online: 25 October 2025
Corresponding author: Raghavendra M. Ichangi
Abstract
A website that is optimized and has a high Search Engine Results Page (SERP) ranking is more likely to attract relevant users. As a result, there is a direct relationship between Search Engine Optimization (SEO) and user experience, and poor SEO makes it hard for a user to find the items he is looking for. The proposed Semantic Self-Attention Regression based on Transformer (SSAR-T) model uses four separate layers—tokenizing, embedding, encoding, and fine-tuning—to determine the degree of user experience satisfaction. Sample input text is fed to the tokenizing layer. The cosine Euclidean semantic similarity-based segment embedding is designed to help minimize the prediction error and training time. Self-attention-based encoder transformation is utilized with multiple attention heads, focusing on learning the context of surrounding words accurately and precisely. Non-linear regression-based fine-tuning is used for measuring customer satisfaction. KANO mapping functions are used to assess the model's precision. Compared to previous methods, the proposed SSAR-T model achieved improvements of 19%, 24% and 58% in precision and 9%, 14% and 12% in recall for SEO, Instagram influencer, and Twitter data samples, respectively.
Keywords:
instagram, twitter, cosine euclidean semantic similarity, self attention-based encoder, non-linear regression, KANO, precision, recallDownloads
References
P. Pookduang, R. Klangbunrueang, W. Chansanam, and T. Lunrasri, ''Advancing Sentiment Analysis: Evaluating RoBERTa against Traditional and Deep Learning Models,'' Engineering, Technology & Applied Science Research, vol. 15, no. 1, pp. 20167–20174, Feb. 2025. DOI: https://doi.org/10.48084/etasr.9703
H. S. Le et al., ''Predictive model for customer satisfaction analytics in E-commerce sector using machine learning and deep learning,'' International Journal of Information Management Data Insights, vol. 4, no. 2, Nov. 2024, Art. no. 100295. DOI: https://doi.org/10.1016/j.jjimei.2024.100295
X. Wang and B. Hu, ''Machine Learning Algorithms for Improved Product Design User Experience,'' IEEE Access, vol. 12, pp. 112810–112821, 2024. DOI: https://doi.org/10.1109/ACCESS.2024.3442085
S. Elnagar, ''Using deep learning to enhance electronic service quality: Application to real estate websites,'' Intelligent Systems with Applications, vol. 21, Mar. 2024, Art. no. 200330. DOI: https://doi.org/10.1016/j.iswa.2024.200330
T. Islam, A. Miron, M. Nandy, J. Choudrie, X. Liu, and Y. Li, ''Transforming Digital Marketing with Generative AI,'' Computers, vol. 13, no. 7, July 2024, Art. no. 168. DOI: https://doi.org/10.3390/computers13070168
Y. H. Ko, P. Y. Hsu, Y. C. Liu, and P. C. Yang, ''Confirming Customer Satisfaction With Tones of Speech,'' IEEE Access, vol. 10, pp. 83236–83248, 2022. DOI: https://doi.org/10.1109/ACCESS.2022.3196733
C. Li, I. Ishak, H. Ibrahim, M. Zolkepli, F. Sidi, and C. Li, ''Deep Learning-Based Recommendation System: Systematic Review and Classification,'' IEEE Access, vol. 11, pp. 113790–113835, 2023. DOI: https://doi.org/10.1109/ACCESS.2023.3323353
M. S. Vimutha and M. C. Padma, ''Insights into Search Engine Optimization using Natural Language Processing and Machine Learning,'' International Journal of Advanced Computer Science and Applications, vol. 14, no. 2, 2023. DOI: https://doi.org/10.14569/IJACSA.2023.0140211
P. Suchanek and N. Bucicova, ''The customer satisfaction model in the mobile telecommunications sector after Covid-19 pandemic,'' PLOS ONE, vol. 20, no. 1, 2025, Art. no. e0317093. DOI: https://doi.org/10.1371/journal.pone.0317093
M. S. Islam et al., ''Challenges and future in deep learning for sentiment analysis: a comprehensive review and a proposed novel hybrid approach,'' Artificial Intelligence Review, vol. 57, no. 3, Mar. 2024, Art. no. 62. DOI: https://doi.org/10.1007/s10462-023-10651-9
M. Ayemowa, R. Ibrahim, and M. M. Khan, ''Analysis of Recommender System Using Generative Artificial Intelligence: A Systematic Literature Review.'' Social Science Research Network, Mar. 02, 2024. DOI: https://doi.org/10.2139/ssrn.4922584
M. A. Kassem, A. A. Abohany, A. A. A. El-Mageed, and K. M. Hosny, ''A novel deep learning model for detection of inconsistency in e-commerce websites,'' Neural Computing and Applications, vol. 36, no. 17, pp. 10339–10353, June 2024. DOI: https://doi.org/10.1007/s00521-024-09590-5
H. J. Tsuei, W. H. Tsai, F. T. Pan, and G. H. Tzeng, ''Improving search engine optimization (SEO) by using hybrid modified MCDM models,'' Artificial Intelligence Review, vol. 53, no. 1, pp. 1–16, Jan. 2020. DOI: https://doi.org/10.1007/s10462-018-9644-0
P. A. Olujimi and A. Ade-Ibijola, ''NLP techniques for automating responses to customer queries: a systematic review,'' Discover Artificial Intelligence, vol. 3, no. 1, May 2023, Art. no. 20. DOI: https://doi.org/10.1007/s44163-023-00065-5
S. Yadav and O. P. Sangwan, ''Neural Network based Approach for Predicting user Satisfaction with Search Engine,'' International Journal of Computer Applications, vol. 18, no. 5, pp. 16–21, Mar. 2011. DOI: https://doi.org/10.5120/2281-2953
"SEO Sample Data." Kaggle, [Online]. Available: https://www.kaggle.com/datasets/muhammetvarl/seo-sample-data.
"Instagram Influencer Dataset." https://sites.google.com/site/sbkimcv/dataset/instagram-influencer-dataset.
A. Goyal, "Twitter-Dataset." Kaggle, [Online]. Available: https://www.kaggle.com/datasets/goyaladi/twitter-dataset.
R. M. Ichangi and S. B. Kulkarni, ''A Discretized Recurrent Deep Learning Classifier based on Stochastic Gradient ChatGPT to Improve Lead Conversion Rate,'' Engineering, Technology & Applied Science Research, vol. 15, no. 3, pp. 22712–22717, June 2025. DOI: https://doi.org/10.48084/etasr.9549
Downloads
How to Cite
License
Copyright (c) 2025 Raghavendra M. Ichangi, Shrinivasrao B. Kulkarni
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
