A deep-learning-based grading system (ASAG) for reading comprehension assessment by using aphorisms as open-answer-questions

Abstract Today reading comprehension is considered an essential skill in modern life, therefore, higher education students require more specific skills to understand, interpret and evaluate texts effectively. Short answer questions (SAQs) are one of the relevant proper tools for assessing skills. Unlike multiple-choice questions, SAQs allow assessment cognitive abilities such as attention, language, perception, problem solving. However, task scoring time-consuming susceptible ambiguity. Automatic Answer Grading (ASAG) a new paradigm that could help solve these problems. This experimental analysis aims implement ASAG using several approaches sentence embedding based on deep learning with multilayer perceptron regression layer top, trained dataset aphorisms. For testing, available composed answers given by 199 undergraduate Spanish. BERT Skip-Thought models tested different hyperparameters find best performance terms Pearson correlation coefficient RMSE against human experts grades. The result current study showed model performed better than other approaches.