ENHANCING SENTIMENT ANALYSIS IN SOCIAL

Abstract

Abstract- Sentiment analysis in social media has gained significant attention due to its wide range of applications in understanding public opinion, market trends, and brand perception. In this paper, we propose and evaluate several deep learning techniques to enhance sentiment analysis accuracy in social media text. We investigate the effectiveness of Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and their variants such as Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs). Our experiments are conducted on a large-scale social media dataset, and we compare the performance of these models in terms of accuracy, precision, recall, and F1-score. Additionally, we explore techniques for handling imbalanced sentiment classes and analyze the impact of different word embeddings on model performance. Our results demonstrate the effectiveness of deep learning approaches in improving sentiment analysis accuracy, especially in the context of noisy and informal social media text. Keywords: Sentiment Analysis, Social Media, Deep Learning, Convolutional Neural Networks, Recurrent Neural Networks, LSTM, GRU, Word Embeddings.