Service Sphere: Citizen Service and Complaint Management System

Globally, particularly in developing countries, registering and tracking citizen complaints remains a significant challenge due to complex government procedures, slow response times, and limited follow-up mechanisms. To address this issue, this study proposes an integrated citizen service and complaint management system that enables users to draft, submit, and monitor complaints through a unified platform. The system leverages fine-tuned large language models (LLMs), including GPT and LLaMA, along with a custom backend that manages routing, microservices, and business logic. The AI models automatically classify complaints and forward them to the appropriate government departments. The platform supports both English and Urdu languages, reducing manual processing and improving accessibility. Experimental results demonstrate that the system achieves 92% classification accuracy, reduces average response time by 40%, and automates 68% of complaint cases. These findings indicate that the proposed system can significantly enhance efficiency, transparency, and citizen satisfaction in public service complaint management.