HCASO: Hybrid Cross-Lingual Attention with Contrastive Alignment and Optimized Lexical Fusion for Bilingual E-Commerce Sentiment Analysis

Himaani Srivatsava; Gangadevi Ganesan

doi:10.7250/csimq.2026-48.03

HCASO: Hybrid Cross-Lingual Attention with Contrastive Alignment and Optimized Lexical Fusion for Bilingual E-Commerce Sentiment Analysis

Authors

Himaani Srivatsava Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai 600089, India https://orcid.org/0009-0002-9431-9962
Gangadevi Ganesan Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai 600089, India https://orcid.org/0000-0003-2339-3989

DOI:

https://doi.org/10.7250/csimq.2026-48.03

Keywords:

Cross-Lingual Sentiment Analysis, Chinese-English Bilingual NLP, Dual-Encoder Transformers, Contrastive Representation Alignment, Gated Lexical Fusion, E-Commerce Review Classification

Abstract

Shared-encoder multilingual models that confound typologically different language representations scarcely support cross-lingual sentiment analysis. This article introduces HCASO (Hybrid Cross-Lingual Attention with Contrastive Alignment and Optimized Lexical Fusion for Sentiment Analysis) as a unified bilingual model, which consists of four elements: (i) language-specific English and Chinese ELECTRA encoders, (ii) cross-lingual attention that allows structured inter-language communication, (iii) a symmetric InfoNCE contrastive loss in a 256-dimensional space that allows the languages to achieve clearer alignment, and (iv) a two-stage gated lexical fusion module that adaptively combines TF-IDF and sentiment lexicon features with contextual encoder representations via a learned per-dimension mixing coefficient. The model is trained using cross-entropy, MSE intensity regression, and the contrastive objectives. Post-hoc temperature scaling was used to calibrate it. Compared to standard baselines, HCASO achieves the strongest overall balance of performance, with ROC-AUC of 0.9361, 0.8623, and 0.8474 on binary, ternary, and quinary MARC English-Chinese sentiment classification, respectively. HCASO performs best on ordinal consistency (MSE = 0.0455, R² = 0.6356, Top-2 = 0.8211) across settings, with XLM-RoBERTa-Base being slightly more accurate in terms of raw accuracy in quinary classification. Ablation experiments with McNemar tests confirmed that all the components - dual encoders (p < 10⁻²⁶ across all settings), lexical fusion (p < 10⁻¹¹ in the ternary setting), and contrastive alignment (p = 0.028 in the quinary setting) each contribute significantly. HCASO is a statistically validated alternative to the shared-encoder paradigm, which scales at 214.4M parameters and 10.89 GMACs.

References

M. K. Nazir, C. N. Faisal, M. A. Habib, and H. Ahmad, “Leveraging Multilingual Transformer for Multiclass Sentiment Analysis in Code-Mixed Data of Low-Resource Languages,” IEEE Access, vol. 13, pp. 7538–7554, 2025.

H. Huang, A. A. Zavareh, and M. B. Mustafa, “Sentiment Analysis in E-Commerce Platforms: A Review of Current Techniques and Future Directions,” IEEE Access, vol. 11, pp. 90367–90382, 2023.

W. Ding, H. Bhuiyan, and M. K. Hossain, “SCOPE: A Transformer-based framework for analyzing sentiment of Chinese public toward delayed retirement,” IEEE Access, vol. 14, pp. 25146–25166, 2026.

E. Kahya Özyirmidokuz, B. Molu Elmas, and E. A. Stoica, “AI-Based Sentiment Analysis of E-Commerce Customer Feedback: A Bilingual Parallel Study on the Fast Food Industry in Turkish and English,” Journal of Theoretical and Applied Electronic Commerce Research, vol. 20, no. 4, pp. 294–294, 2025.

M. F. Almufareh, N. Jhanjhi, N. A. Khan, S. N. Almuayqil, M. Humayun, and D. Javed, “BertSent: Transformer-Based Model for Sentiment Analysis of Penta-Class Tweet Classification,” IEEE Access, vol. 12, pp. 196803–196817, 2024.

S. Shan, J. Sun, and R. M. C. Macawile, “Examining Customer Satisfaction through Transformer-Based Sentiment Analysis for Improving Bilingual E-Commerce Experiences,” IEEE Access, vol. 13, pp. 51027–51041, 2025.

Y. Zhang, Q. Ma, Y. Zhang, and Y. Liu, “Sentiment Analysis for E-Commerce Product Reviews Using a Fourier Graph Transformer Network,” IEEE Access, vol. 13, pp. 194127–194138, 2025.

A. Ishtiaq, K. Munir, A. Raza, N. A. Samee, M. M. Jamjoom, and Z. Ullah, “Product Helpfulness Detection With Novel Transformer Based BERT Embedding and Class Probability Features,” IEEE Access, vol. 12, pp. 55905–55917, 2024.

F. Ullah et al., “Prompt-based fine-tuning with multilingual transformers for language-independent sentiment analysis,” Scientific Reports, vol. 15, article 20834, 2025.

I. Duru and A. S. Sunar, “Transformer and Pre-Transformer Model-Based Sentiment Prediction with Various Embeddings: A Case Study on Amazon Reviews,” Entropy, vol. 27, no. 12, pp. 1202–1202, 2025.

V. Naveen and S. Sountharrajan, “A Lightweight Multimodal Framework with Dilated CNN Audio Modelling and SE-Enhanced MobileNetV2 for Real-Time Sentiment Analysis of Customer Product Reviews,” IEEE Access, vol. 13, pp. 217825–217845, 2025.

N. Raychawdhary, S. Bhattacharya, C. Seals, and G. Dozier, “Empowering Sentiment Analysis in African Low-Resource Languages Through Transformer Models and Strategic Language Selection,” IEEE Access, vol. 13, pp. 147859–147873, 2025.

N. Kai Ning Loh, C. Poo Lee, T. Song Ong and K. Ming Lim, “MPNet-GRUs: Sentiment Analysis With Masked and Permuted Pre-Training for Language Understanding and Gated Recurrent Units,” IEEE Access, vol. 12, pp. 74069–74080, 2024.

M. Akash Rahman, M. Begum, T. Mahmud, M. S. Hossain, and K. Andersson, “Analyzing Sentiments in eLearning: A Comparative Study of Bangla and Romanized Bangla Text Using Transformers,” IEEE Access, vol. 12, pp. 89144–89162, 2024.

E. Hashmi, S. Y. Yayilgan, and S. Shaikh, “Augmenting sentiment prediction capabilities for code-mixed tweets with multilingual transformers,” Social Network Analysis and Mining, vol. 14, article 86, 2024.

K. R. Mabokela, T. Celik, and M. Raborife, “Multilingual Sentiment Analysis for Under-Resourced Languages: A Systematic Review of the Landscape,” IEEE Access, vol. 11, pp. 15996–16020, 2023.

M.S.U. Miah, M. M. Kabir, T. B. Sarwar, M. Safran, S. Alfarhood, and M. F. Mridha, “A multimodal approach to cross-lingual sentiment analysis with ensemble of transformer and LLM,” Scientific Reports, vol. 14, article 9603, 2024.

G. Mu, C. Chen, X. Li, J. Li, X. Ju, and J. Dai, “Multimodal Sentiment Analysis of Government Information Comments Based on Contrastive Learning and Cross-Attention Fusion Networks,” IEEE Access, vol. 12, pp. 165525–165538, 2024.

S. A. M. Zaidi, S. Latif, and J. Qadir, “Enhancing Cross-Language Multimodal Emotion Recognition With Dual Attention Transformers,” IEEE Open Journal of the Computer Society, vol. 5, pp. 684–693, 2024.

C. Peng, C. Zhang, X. Xue, J. Gao, H. Liang, and Z. Niu, “Cross-modal complementary network with hierarchical fusion for multimodal sentiment classification,” Tsinghua Science and Technology, vol. 27, no. 4, pp. 664–679, 2022.

K. Zhao, M. Zheng, Q. Li, and J. Liu, “Multimodal Sentiment Analysis – A Comprehensive Survey From a Fusion Methods Perspective,” IEEE Access, vol. 13, pp. 64556–64583, 2025.

K. R. Jerripothula, A. Rai, K. Garg, and Y. S. Rautela, “Feature-Level Rating System Using Customer Reviews and Review Votes,” IEEE Transactions on Computational Social Systems, vol. 7, no. 5, pp. 1210–1219, 2020.

S. M. Dashtaki, M. H. Chagahi, A. Bahadori, B. Moshiri, Md. J. Piran, and A. Montazeri, “HSIF: A Transformer-Based Cross-Attention Framework for Cryptocurrency Trend Forecasting via Multimodal Sentiment–Market Fusion,” IEEE Access, vol. 13, pp. 156600–156612, 2025.

H. Fan and J. Chen, “Position Perceptive Multi-Hop Fusion Network for Multimodal Aspect-Based Sentiment Analysis,” IEEE Access, vol. 12, pp. 90586–90595, 2024.

Y. Zhang, C. Zhu, and Y. Xie, “Fine-Grained Sentiment Analysis of Cross-Domain Chinese E-Commerce Texts Based on SKEP_Gram-CDNN,” IEEE Access, vol. 11, pp. 74058–74070, 2023.

A. V. Kotelnikova, S. V. Vychegzhanin, and E. V. Kotelnikov, “Cross-Domain Sentiment Analysis Based on Small in-Domain Fine-Tuning,” IEEE Access, vol. 11, pp. 41061–41074, 2023.

S. Nayab, M. K. Hanif, R. Talib, and M. U. Sarwar, “Aspect-Context Level Information Extraction via Transformer Based Interactive Attention Mechanism for Sentiment Classification,” IEEE Access, vol. 11, pp. 57683–57692, 2023.

M. R. R. Rana, A. Nawaz, S. U. Rehman, M. A. Abid, M. Garayevi, and J. Kajanová, “BERT-BiGRU-Senti-GCN: An Advanced NLP Framework for Analyzing Customer Sentiments in E-Commerce,” International Journal of Computational Intelligence Systems, vol. 18, article 21, 2025.

E. Hashmi and S. Y. Yayilgan, “A robust hybrid approach with product context-aware learning and explainable AI for sentiment analysis in Amazon user reviews,” Electronic Commerce Research, vol. 25, pp. 5139–5171, 2025.

P. Keung, Y. Lu, G. Szarvas, and N. A. Smith, “The Multilingual Amazon Reviews Corpus,” in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020, pp. 4563–4568.

P. Bhargava, A. Drozd, and A. Rogers, “Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristics,” in Proceedings of the Second Workshop on Insights from Negative Results in NLP, 2021, pp. 125–135.

M. Ni, Z. Sun, and W. Liu, “Reversible jump attack to textual classifiers with modification reduction,” Machine Learning, vol. 113, pp. 5907–5937, 2024.

C. Guo, G. Pleiss, Y. Sun, and K. Weinberger, “On Calibration of Modern Neural Networks,” in Proceedings of the 34th International Conference on Machine Learning, PMLR, 2017, pp. 1321–1330.