Multilingual Sentimental Analysis of You Tube Comments

The increasing use of multilingual platforms like YouTube has created a demand for effective sentiment analysis tools capable of processing diverse languages. This paper presents a comprehensive approach to multilingual sentiment analysis, focusing on English, Kannada, Hindi, Telugu comments. The study incorporates advanced Natural Language Processing (NLP) techniques, such as tokenization, stemming, and TF-IDF vectorization, alongside machine learning algorithms like Logistic Regression, Random Forest, and Linear SVC. The system employs a modular architecture, including language detection, feature extraction, and sentiment classification. Results demonstrate the model's ability to identify emotions with high accuracy, contributing to content creators' and marketers' efforts to understand audience engagement. Challenges like code-switching and limited language resources are discussed, with potential enhancements proposed to improve inclusivity and scalability.