A Hybrid Automated Essay Scoring Using NLP and Random Forest Regression
Muhammad Zaim Azri Bin Azahar, Khairil Imran Ghauth · 2022
Assessing the performance of students through subjective assessments namely essays is critical in measuring their achievement during the learning process in an educational system.The essay test will evaluate the student's ability to remember and express their ideas or opinions toward certain topics.A teaching staff is usually required to assess and grade the students' essays.This paper presents a hybrid Automated Scoring System based on Natural Language Processing (NLP) and Random Forest Regression.The model focused on regression task where the predicted score is in a continuous value.Natural Language Processing (NLP) has also been applied in this work to extract features from essays.Finally, all the proposed model is compared to Linear Regression and Deep Learning and are then evaluated to compare the performance of the models by using the Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE).