Context-Aware Sentiment Classification of Movie Reviews Using Bidirectional LSTM Networks
International Journal of Scientific Research in Science, Engineering and Technology · 25.03.2026 · 13 (2) · 159-171
DOI: 10.32628/ijsrset261371 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Scientific Research in Science, Engineering and Technology, online erschienen am 25.03.2026. Verfasst von Pandiri Lavanya und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The rapid growth of digital platforms has led to a massive increase in user-generated textual data, especially in the form of online movie reviews. Analyzing such unstructured data manually is inefficient and time-consuming. Sentiment Analysis, a key application of Natural Language Processing (NLP), enables automatic classification of opinions expressed in text. This study proposes a deep learning-based approach to classify movie reviews into positive and negative sentiments. The IMDB Movie Review Dataset, containing 50,000 labeled reviews, was used for this research. Data preprocessing techniques such as HTML tag removal, text normalization, tokenization, stop-word removal, and stemming were applied to improve data quality. Word embedding methods were used to convert text into numerical vectors, allowing the model to capture semantic relationships between words. A Bidirectional Long Short-Term Memory (BiLSTM) model was implemented for sentiment classification. Unlike traditional machine learning models, BiLSTM processes text in both forward and backward directions, enabling better understanding of context and sentence structure. The model was trained using the Adam optimizer and Binary Cross-Entropy loss function, with regularization techniques applied to prevent overfitting. The proposed model achieved a validation accuracy of approximately 88.53%, demonstrating its effectiveness in sentiment classification. The results indicate that deep learning approaches, particularly BiLSTM, outperform traditional methods in handling sequential and context-dependent text data. This system can be applied in real-world scenarios such as movie recommendation systems, social media analysis, and customer feedback evaluation. Future work includes integrating transformer-based models like BERT and extending the system to multilingual datasets.
Zitieren
APA
Lavanya, P., Prasad, P. D., & Duvvuri, S. K. (2026). Context-Aware Sentiment Classification of Movie Reviews Using Bidirectional LSTM Networks. International Journal of Scientific Research in Science, Engineering and Technology, 13(2), 159-171. https://doi.org/10.32628/ijsrset261371
Vancouver
Lavanya P, Prasad PD, Duvvuri SK. Context-Aware Sentiment Classification of Movie Reviews Using Bidirectional LSTM Networks. International Journal of Scientific Research in Science, Engineering and Technology. 2026;13(2):159-171. doi:10.32628/ijsrset261371
BibTeX
@article{Lavanya2026,
author = {Lavanya, Pandiri and Prasad, Patinavalasa Durga and Duvvuri, Suneel Kumar},
title = {Context-Aware Sentiment Classification of Movie Reviews Using Bidirectional LSTM Networks},
journal = {International Journal of Scientific Research in Science, Engineering and Technology},
year = {2026},
volume = {13},
number = {2},
pages = {159--171},
doi = {10.32628/ijsrset261371}
}
EUVIMED