Twitter Sentiment Analysis using a Hybrid Model of Deep Recurrent Neural Networks with Graph Convolutional Networks

The quantitative investigation of opinions, sentiments, and emotions as they are expressed in written language is called sentiment analysis, sometimes referred to as opinion mining. In recent years, text mining and natural language processing (NLP) have emerged as two of the most prominent research areas in this domain. The appeal of sentiment analysis lies in two key aspects: first, opinions are central to nearly all human activities and serve as powerful drivers of decision-making; second, in the face of uncertainty, individuals often rely on the perspectives of others to guide their choices. Deep Learning (DL) has gained significant attention for its exceptional performance across a wide range of applications. In this study, we propose a hybrid deep learning model that integrates Recurrent Neural Networks (RNNs) with Graph Convolutional Networks (GCNs) to analyze the sentiment of Twitter reviews. The RNN component effectively captures the sequential and contextual dependencies of textual data, while the GCN component leverages relational information and word co-occurrence structures to enhance feature representation. This hybrid approach aims to improve sentiment classification accuracy performance of 94.75% by integrating both temporal and structural information from tweets.