Published Paper Details:

In order to predict election outcomes, we are using the power of large language models, particularly GPT, to conduct sentiment analysis on Twitter. The extensive use of digital technology has led to a significant increase in the production of user-generated material, which has in turn sparked a radical change in the dynamics of communication across various platforms. In particular, social media platforms have become treasure troves of behavioral data that provide deep insights into a variety of fields, such as politics, e-commerce, medicine, and education. Predictive analytics pertaining to political tweet mining poses significant challenges, chief among them being the accurate assessment of sentiment accuracy and the detection of propagandistic narratives. Due to LLMs' proficiency in natural language processing (NLP) tasks, we suggest using them as a solution, especially GPT. Due to their thorough training, LLMs are able to comprehend sentiment and context, among other complex linguistic nuances. Their ability to generate cohesive text is essential for sentiment analysis. By utilizing these benefits, our goal is to use sentiment analysis with GPT models to forecast the results of the Indian Lok Sabha Elections in 2024. This study leverages the capabilities of LLMs and NLP approaches to solve the urgent demand for credible methodologies in election outcome prediction.