Exploring the Efficacy of ChatGPT in Adapting Reading Materials for Undergraduate Students

Abstract

[EN] Current research explores the efficiency of ChatGPT 3.5 in text adaptation for educational purposes. It aims to investigate reliability of the AI model for reading ease evaluation, its accuracy in reading manipulation on specific parameters, output quality, and prospects of improving user experience. The study considered text, sentence, and word length as the main features for measuring text difficulty. Flesch-Kincaid reading ease statistics was used for input and output text accessibility evaluation. The research identifies the areas of concern of using ChatGPT 3.5 for text manipulation and requirements for successful implementation of the AI model in the process of the reading materials transformation.