Comprehensive Assessment and Optimization of Sentiment Analysis Models for Movie Reviews with Enhanced Movie Recommendation Systems

Sentiment analysis is the most important process for considering public sentiment towards movies. This study estimates classifiers using TMDb movie reviews, which identify effective sentiment analysis methods. Here, Innumerable classifiers are compared to optimize sentiment classification accuracy. Here, TMDb API is used to fetch movie reviews stored after preprocessing. Classifiers such as Naive Bayes, SVMs, and MLP are evaluated with TfidfVectorizer for text processing. Then, Performance metrics like accuracy and ROC-AUC scores are designed with SVM. Linear SVC proved to be the best classifier, excelling in evaluation measures such as F1-score, accuracy, and ROC-AUC. SVM performance is boosted through GridSearchCV by optimal parameter tuning, which represents vigorous sentiment analysis capability for movie reviews. Automated sentiment analysis is advantageous for critics and review platforms, improving review processing efficiency. The scalable methodology boosts decision-making across industries for marketing and audience engagement strategies for filmmakers. The challenges triggered in movie recommendation systems were reported by a novel approach introduced in this paper, which combines Count Vectorization, Cosine Similarity, and Truncated Singular Value Decomposition (SVD) by utilizing practices likewise Linear SVC for sentiment analysis and Linear Regression for rating prediction by using Hyperparameter tuning and comparison with basic method implementation, This Study exhibited the importance of fine-tuning models for better accuracy. Our method effectively reveals issues like data sparsity and cold start via extensive investigation and evaluation. Our methodology offers a new era of better recommendation accuracy and effective sentiment analysis, improving movie recommendation systems. Ultimately, this research introduces a novel approach that combines sentiment analysis, hyperparameter tuning, and feature scaling to augment movie recommendation accuracy, distinguishing it from present studies.