Indonesian automatic short answer grading system

Short answer question is one of the methods used to evaluate student cognitive abilities, including memorizing, designing, and freely expressing answers based on their thoughts. Unfortunately, grading short more complicated than multiple choices answers. For that problem, several studies have tried build an artificial intelligence system called automatic (ASAG). We improve accuracy ASAG at scoring in Indonesian by enhancing earlier state-of-the-art models methods. They were bidirectional encoder representations from transformer (BERT) with fine-tuning approach ridge regression utilizing advanced feature extraction. conducted this study doing stages literature review, data set preparation, model development, implementation, comparison. Using two different sets, best result was achievement 0.9508 pearson’s correlation 0.4138 root-mean-square error (RMSE) BERT-based approach. This outperformed results previous using same evaluation metrics. Thus, it proved our BERT can