ElectionBot for Voters: Bridging the Gap Between Data and Decision-Making with LLM

Voters often do not have sufficient access to reliable candidate information during election periods, leaving them susceptible to misinformation from unreliable sources. And even if information is available, it is often challenging to search through it because of its overwhelming volume, making it nearly unreadable. The proposed question-answering model for elections aims to solve all these problems. It will provide all the information necessary to make an educated decision and choose the right candidate on a single platform. Using a Large Language Model, we have aimed to create a platform where users can effort-lessly inquire about upcoming election candidates using plain English. The model's proficiency lies in its prompt-engineered architecture coupled with a zero-shot methodology. This robust approach showcases the model's adeptness in understanding diverse user queries in natural language and responding with accurate answers regarding the polls. This paper encompasses all the research that helped determine the optimal model, as well as the approach, findings, and results.