Revolutionizing Assessment: AI-Powered Evaluation with RAG and LLM Technologies

The world of Artificial Intelligence is rapidly evolving after the introduction of GEN AI. Artificial Intelligence is being adopted in many fields and to automate complex works which frees up humans to something better. In the field of education, Artificial Intelligence is used in various ways, to tailor the content to meet the needs of individual students, one to one tutoring experience, as teaching assistant, in planning, content development. The scope AI and its subfield can be leveraged further, and it is possible to use it for exam assessment, use the data from Student Management System to generate some meaningful insight on Students behavior, discipline, attendance etc. Natural Language Processing, a subset of Artificial Intelligence can be combined with other AI subsets like Large Language Model to facilitate the Teacher, Parents and Students interaction, Parents can use chatBots to receive candid feedback of their kid’s performance, to generate relative performance improvement compared to previous cycle, suggestion for improvement. All it takes is to connect the Artificial Intelligence subsets with some reference system like Students database. This research analysis is to evaluate student’s performance and provide feedback factoring based on four main areas like exam answer sheet, practicals, attendance and assignments. This research provides technical implementation and guidance to implement the solution using the Retrieval Augmented Generation and Large Language Model, and an architectural overview of combining data from various input modes like student information from Student Management System, text book, digital answer sheet and embedding them in the Vector database.