Questions Classification According to Bloom’s Taxonomy using Universal Dependency and Word Net

Abstract

Question Classification (QC) based on Bloom’s Taxonomy has been widely accepted and used as a guideline in designing a holistic set of examination questions that consists of various cognitive levels. However, many discrepancies happened in QC due to inconsistence or misclassification of questions to Bloom’slevel. This paper proposes a system that can analyze the examination questions and determine the appropriate Bloom’s levels using syntactical and semantic approach. Universal Dependency (UD) that implies Natural Language Processing (NLP) technique is used to identify the important keywords and verbs. Then, WordNet similarity algorithm with Natural Language Toolkit (NLTK) is used to identify the questions category according to the Bloom’s Taxonomy. This research focuses on Science, Technology, Engineering, and Mathematics (STEM) examination questions. At present, a set of 100 questions is used and preliminary result indicates both Universal Dependency and WordNet similarity algorithms being able to categorize successfully the questions based on Bloom’s Taxonomy.