Answer Categorization Method Using K-Means for Indonesian Language Automatic Short Answer Grading System Based on Latent Semantic Analysis

The Automatic Short Answer Grading (Simple-O) has been created for grading short answer with Bahasa Indonesia using K-Means and Latent Semantic Analysis (LSA) method. In this experiment, the text document feature will be extracted using Term Frequency-Inverse Document Frequency (TF-IDF) and then classified using K-Means. From the experiment, 149 documents are expected to be clustered into five classes. The result of the clustering using K-Means is 100% matched with clustering using human rater. The result of grading with LSA is 74%.