HYBRID EXPLAINABLE SENTIMENT ANALYSIS FRAMEWORK

Vidyadhari S; Professor B. Prajna

doi:10.5281/zenodo.22726839

Abstract

Sentiment analysis is widely used to identify opinions expressed in user-generated text, but conventional singlemodel approaches may provide limited robustness and limited interpretability. This paper presents a hybrid and
explainable aspect-based sentiment analysis framework that integrates TF-IDF feature representation, a Linear
Support Vector Machine (SVM), VADER sentiment analysis, DistilBERT-based sentiment inference, confidence
scoring, SHAP-based explanation, and an interactive Streamlit dashboard. The system processes restaurantdomain aspect annotations from the SemEval-2014 Task 4 ABSA dataset. Each instance combines the target
aspect and sentence context using an aspect-aware representation. TF-IDF with unigram and bigram features
forms the primary supervised representation, while VADER and DistilBERT provide complementary sentiment
signals. A weighted fusion mechanism gives the major contribution to the SVM prediction and smaller
contributions to the other models. SHAP is used as a post-hoc explanation layer for the linear model, while the
dashboard presents aspect-level sentiment, confidence information, analytics, word clouds, and downloadable
results. The reported project evaluation obtained 63.69% for VADER and 93.28% for TF-IDF with SVM, with
the hybrid configuration also reported at 93.28%. These figures are presented as reported project evaluation results
rather than independent held-out test performance. The framework demonstrates an end-to-end practical approach
for combining prediction, confidence information, explainability, and visualization in restaurant-review analysis.

Keywords

Aspect-Based Sentiment Analysis, TF-IDF, SVM, VADER, SHAP, Explainable AI

References

1) M. Pontiki, D. Galanis, J. Pavlopoulos, H. Papageorgiou, I. Androutsopoulos, and S. Manandhar,
“SemEval-2014 Task 4: Aspect Based Sentiment Analysis,” in Proceedings of the 8th International
Workshop on Semantic Evaluation (SemEval 2014), pp. 27–35, 2014. doi: 10.3115/v1/S14-2004.
2) C. J. Hutto and E. Gilbert, “VADER: A Parsimonious Rule-Based Model for Sentiment Analysis of
Social Media Text,” Proceedings of the International AAAI Conference on Web and Social Media, vol. 8,
no. 1, pp. 216–225, 2014. doi: 10.1609/icwsm.v8i1.14550.
3) C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273–297, 1995.
doi: 10.1007/BF00994018.
4) T. Joachims, “Text Categorization with Support Vector Machines: Learning with Many Relevant
Features,” in Machine Learning: ECML-98, pp. 137–142, Springer, 1998. doi: 10.1007/BFB0026683.
5) J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “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,
pp. 4171–4186, 2019. doi: 10.18653/v1/N19-1423.
6) V. Sanh, L. Debut, J. Chaumond, and T. Wolf, “DistilBERT, a distilled version of BERT: smaller, faster,
cheaper and lighter,” arXiv preprint arXiv:1910.01108, 2019.
7) S. M. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” in Advances in
Neural Information Processing Systems 30 (NeurIPS 2017), pp. 4765–4774, 2017.
8) M. T. Ribeiro, S. Singh, and C. Guestrin, “Why Should I Trust You? Explaining the Predictions of Any
Classifier,” in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge
Discovery and Data Mining, pp. 1135–1144, 2016. doi: 10.1145/2939672.2939778.
9) F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer,
R. Weiss, V. Dubourg, J. Vanderplas, J. Passos, D. Cournapeau, M. Brucher, M. Perrot, and É.
Duchesnay, “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12,
pp. 2825–2830, 2011.
10) W. McKinney, Python for Data Analysis: Data Wrangling with Pandas, NumPy, and IPython, 2nd
ed. Sebastopol, CA, USA: O’Reilly Media, 2017.
11) S. Raschka, Y. Liu, and V. Mirjalili, Machine Learning with PyTorch and Scikit-Learn: Develop
Machine Learning and Deep Learning Models with Python. Birmingham, UK: Packt Publishing, 2022.
12) Streamlit, “Streamlit Documentation.” [Online]. Available: https://docs.streamlit.io/