Efficient Redundancy Detection in Large-Scale Examinations: A SRNN Model for Enhanced Test Quality and Resource Optimization

Abstract

This project aims to develop an artificial intelligence model capable of efficiently detecting and eliminating redundancies in multiple-choice tests used in large-scale competitive examinations. By exploiting the data processing capabilities of deep learning, this model aims to improve the quality and reliability of the tests. This will ensure a fairer and more equitable assessment of candidates' skills but also optimize the time and resources needed to design and revise items. The project aims to facilitate the work of test designers by automating part of the revision process while maintaining high standards of item diversity and relevance.