International Journal of Scientific Research and Engineering Development

The rapid increase in the number of global visa applications has added a lot of complexities to immigration decision making, which made the traditional manual process both lengthy and inconsistent. In order to overcome such difficulties, the present paper proposes an AI-enabled Visa Approval Prediction System that utilizes machine learning together with MLOps to facilitate the process of making correct, efficient and scalable immigration decisions. The proposed framework involves the analysis of historical visa application data through the consideration of such attributes of applicants as education, experience, prevailing wage, company, employment region and job details. The selection of machine learning algorithm that provides the best possible prediction results out of several supervised machine learning algorithms is carried out. The whole process of machine learning is fully automated using an MLOps pipeline consisting of the stages of data ingestion, preprocessing, training, validation, deployment and monitoring. The proposed framework relies on Docker for containerization, GitHub Actions for CI/CD pipeline automation, AWS for the scalable deployment and MongoDB for the management of the data efficiently. Continuous monitoringand model versioning help to keep the performance of the model consistent.

Assistant Prof.Sentil Kumar V, Afreed Ahmed, Ashok Gowda SP, Keerthana G, "AI Driven Immigration Decision Support System" International Journal of Scientific Research and Engineering Development, V9(4): Page(960-965) July-August 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.