Author(s) :

Mr.Y.Srikar.

Volume/Issue :

Abstract :

This paper presents a comparative analysis of sentiment analysis methodologies on social media using advanced deep learning techniques, achieving an accuracy of 98%. The study introduces a comprehensive framework that addresses all phases of the deep learning process, including data collection, preprocessing, feature extraction, and model optimization. It employs Bidirectional Long Short-Term Memory networks (BiLSTMs) to capture contextual information and utilizes feature extraction methods such as TF-IDF, Word2Vec,and GloVe for effective text representation. The methodology incorporates Scikit-learn and Gensim for hyperparameter tuning and feature extraction, respectively, with GridSearchCV optimizing model parameters.The results demonstrate that this approach significantly surpasses existing techniques, setting new benchmarks in sentiment analysis accuracy. By comparing these methods with prior studies, the research highlights substantial improvements in model robustness and generalizability, providing critical insights for advancing sentiment analysis in social media. period.

No. of Downloads :

0