Development of a tool to support exam evaluation using OCR and LLMs

Identifier:  TFG:9459

Handle: https://hdl.handle.net/20.500.11797/TFG9459

Authors:  Soriano Frutos, Luis Miguel

Abstract:
This work aims to develop a support tool for the correction of handwritten exams by combining optical character recognition models and large language models. In addition, a comparative analysis is carried out between different character recognition models to determine which one offers greater accuracy. This system seeks to facilitate and optimize the correction process and reduce the workload of teachers, thus improving the efficiency of academic assessment. This approach allows to assist a significant part of the educational process without replacing the intervention of the teacher or compromising the quality of the assessment.