Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Sentiment Analysis on Movie Reviews: A Deep Dive into Modern Techniques and Open Challenges

Agnivo Gosai, Shuvodeep De, Karun Thankachan

TL;DR

The paper surveys sentiment analysis on movie reviews, tracing the evolution from rule-based and lexicon-driven methods to deep learning, transformers, and large language models. It synthesizes benchmark datasets and evaluation frameworks, highlighting a shift from binary accuracy to multimodal and fine-grained emotion analyses, while critically examining benchmark limitations. Five core challenges—sarcasm/irony, domain and temporal drift, long-form context, interpretability, and resource efficiency—are analyzed for their impact on real-world deployment. The authors propose a roadmap emphasizing improved benchmarks, cross-lingual transfer, explainable and fair models, multimodal fusion, and practical deployment considerations for scalable, robust sentiment systems in the movie domain.

Abstract

This paper presents a comprehensive survey of sentiment analysis methods for movie reviews, a benchmark task that has played a central role in advancing natural language processing. We review the evolution of techniques from early lexicon-based and classical machine learning approaches to modern deep learning architectures and large language models, covering widely used datasets such as IMDb, Rotten Tomatoes, and SST-2, and models ranging from Naive Bayes and support vector machines to LSTM networks, BERT, and attention-based transformers. Beyond summarizing prior work, this survey differentiates itself by offering a comparative, challenge-driven analysis of how these modeling paradigms address domain-specific issues such as sarcasm, negation, contextual ambiguity, and domain shift, which remain open problems in existing literature. Unlike earlier reviews that focus primarily on text-only pipelines, we also synthesize recent advances in multimodal sentiment analysis that integrate textual, audio, and visual cues from movie trailers and clips. In addition, we examine emerging concerns related to interpretability, fairness, and robustness that are often underexplored in prior surveys, and we outline future research directions including zero-shot and few-shot learning, hybrid symbolic--neural models, and real-time deployment considerations. Overall, this abstract provides a domain-focused roadmap that highlights both established solutions and unresolved challenges toward building more accurate, generalizable, and explainable sentiment analysis systems for movie review data.

Sentiment Analysis on Movie Reviews: A Deep Dive into Modern Techniques and Open Challenges

TL;DR

The paper surveys sentiment analysis on movie reviews, tracing the evolution from rule-based and lexicon-driven methods to deep learning, transformers, and large language models. It synthesizes benchmark datasets and evaluation frameworks, highlighting a shift from binary accuracy to multimodal and fine-grained emotion analyses, while critically examining benchmark limitations. Five core challenges—sarcasm/irony, domain and temporal drift, long-form context, interpretability, and resource efficiency—are analyzed for their impact on real-world deployment. The authors propose a roadmap emphasizing improved benchmarks, cross-lingual transfer, explainable and fair models, multimodal fusion, and practical deployment considerations for scalable, robust sentiment systems in the movie domain.

Abstract

This paper presents a comprehensive survey of sentiment analysis methods for movie reviews, a benchmark task that has played a central role in advancing natural language processing. We review the evolution of techniques from early lexicon-based and classical machine learning approaches to modern deep learning architectures and large language models, covering widely used datasets such as IMDb, Rotten Tomatoes, and SST-2, and models ranging from Naive Bayes and support vector machines to LSTM networks, BERT, and attention-based transformers. Beyond summarizing prior work, this survey differentiates itself by offering a comparative, challenge-driven analysis of how these modeling paradigms address domain-specific issues such as sarcasm, negation, contextual ambiguity, and domain shift, which remain open problems in existing literature. Unlike earlier reviews that focus primarily on text-only pipelines, we also synthesize recent advances in multimodal sentiment analysis that integrate textual, audio, and visual cues from movie trailers and clips. In addition, we examine emerging concerns related to interpretability, fairness, and robustness that are often underexplored in prior surveys, and we outline future research directions including zero-shot and few-shot learning, hybrid symbolic--neural models, and real-time deployment considerations. Overall, this abstract provides a domain-focused roadmap that highlights both established solutions and unresolved challenges toward building more accurate, generalizable, and explainable sentiment analysis systems for movie review data.
Paper Structure (72 sections, 3 figures, 3 tables)

This paper contains 72 sections, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Significance and Motivation
  4. The Evolution of Sentiment Analysis
  5. Limitations and Challenges
  6. Novelty and Scope of this Review
  7. Problem Formulation and Taxonomy
  8. Granularity of Sentiment Analysis
  9. Types of Movie Review Data
  10. Unstructured Textual Reviews
  11. Structured Ratings and Metadata
  12. Annotated Corpora for Fine-Grained Analysis
  13. Benchmark Datasets and Metrics
  14. Foundational Datasets: Establishing the Benchmark Paradigm
  15. The Movie Reviews Corpus: Pioneering Binary Classification
  16. ...and 57 more sections

Figures (3)

  • Figure 1: Sentiment Analysis on Movie Review: Organization of the Article. Figure created with AI assistance.
  • Figure 2: Timeline of major benchmark dataset development in movie review sentiment analysis, showing the evolution from simple binary classification to complex multimodal evaluation frameworks.
  • Figure 3: The progression of peak accuracy on IMDB benchmarks versus growing awareness of persistent challenges. As accuracy approaches saturation, attention shifts to robustness, explainability, and deployment considerations.