Questions Classification Software based on Bloom’s Cognitive Levels using Naive Bayes Classifier Method

Abstract

Questions Classification is one way to know how the student understanding some lessons. Those questions can be collected and classified based on cognitive Bloom level. Bloom Cognitive Level organized question in groups that represents contents of those questions. Words contained in every question have certain meaning and can be used as base to determine category of question. This study aims to classify amounts of questions based on cognitive Bloom level with Naive Bayes Classifier method. According to Bloom's taxonomy of cognitive level divided into six levels of the Knowledge (C1), Comprehension (C2), Application (C3), Analysis (C4), Synthesis (C5), and Evaluation (C6). In this study, questions classified into 3 classes based on cognitive Bloom level, Knowledge (C1), Comprehension (C2), Application (C3). The amount of collective data used for training process is 240 questions. Result of this study generates accuracy of 75 % from 60 question samples tested. Susceptibility often occured because of same vocabularies from each categories, thus cause mistakes in classification.