2026
doi: 10.12720/jait.17.9.1821-1830
Sentiment Classification on IMDb Movie Reviews: A Controlled Comparative Study of Classical, Deep-learning, Feature-fusion, and Fine-tuned Transformer Models
Santosh Kumar Banbhrani1,2,3,*
,
Wang Shaojie1,* ,
Liu Baiyang1 ,
Chen Wei1 ,
Pir Dino Soomro4 ,
Ali Raza Rang2,3 ,
Dhani Bux Talpur2,3 ,
Asad Ullah Shah3
- 1School of Electrical Engineering, Shaoyang University, Shaoyang, China
- 2Department of Information and Computing, University of Sufism and Modern Science, Bhitshah, Matiari, Sindh, Pakistan
- 3Kulliyyah of Information and Communication Technology, International Islamic University Malaysia, Jalan Gombak, Kuala Lumpur, Selangor, Malaysia
- 4School of Information Science and Technology, Dalian Maritime University, Dalian, China
- Manuscript receivedFebruary 23, 2026
- revisedMarch 10, 2026
- acceptedJuly 10, 2026
- publishedSeptember 24, 2026
Abstract
Sentiment analysis is one of the most important tasks in natural language processing, as it allows opinions expressed in text to be automatically categorized. This study presents a systematically controlled comparative analysis of classical machine-learning, deep-learning, feature-fusion, and fully fine-tuned transformer models for binary sentiment classification on the Internet Movie Database (IMDb) Movie Reviews dataset. While the individual components of the framework, including Term Frequency-Inverse Document Frequency (TF-IDF), Bidirectional Encoder Representations from Transformers (BERT) embeddings, feature fusion, and Multi-Layer Perceptron (MLP) classification, are established techniques, this study contributes a comprehensive empirical evaluation of their combined effectiveness for the target classification task, providing insights into the practical benefits of integrating lexical and contextual representations. Under a single experimental framework, all models are evaluated on a commonly held-out test set with 95% bootstrap confidence intervals, paired McNemar significance testing, confusion-matrix and error analysis, and a computational-complexity comparison. The fully fine-tuned transformers achieve the highest performance Robustly Optimized BERT Pretraining Approach (RoBERTa), F1-Score = 0.9427. The TF-IDF and BERT feature-fusion model is significantly more accurate than the classical, deep-learning, and BERT-only baselines (all p < 0.001), but significantly less accurate than the fine-tuned transformers (all p < 0.01); furthermore, concatenating the [CLS] token and mean-pooled representations does not significantly improve upon mean pooling alone ( p = 0.48). The study provides a transparent, evidence-based account of the accuracy and cost trade-offs among these model families.
Keywords
- sentiment analysis
- Internet Movie Database (IMDb) dataset
- machine learning
- deep learning
- hybrid models
- Term Frequency-Inverse Document Frequency (TF-IDF)
- Bidirectional Encoder Representations from Transformers (BERT)
How to Cite
Santosh Kumar Banbhrani, Wang Shaojie, Liu Baiyang, Chen Wei, Pir Dino Soomro, Ali Raza Rang, Dhani Bux Talpur, and Asad Ullah Shah, "Sentiment Classification on IMDb Movie Reviews: A Controlled Comparative Study of Classical, Deep-learning, Feature-fusion, and Fine-tuned Transformer Models," Journal of Advances in Information Technology, vol. 17, no. 9, pp. 1821-1830, 2026. https://doi.org/10.12720/jait.17.9.1821-1830
Copyright & License
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
