Parts of speech in Bloom’s Taxonomy Classification

Abstract

This paper analyses parts of speech in a training corpus with 13,189 learning outcomes in which Bloom’s Taxonomy levels were previously classified by human experts for 3,496 subjects offered at an Australian university. This paper explores the automatic identification of verbs and other parts of speech impacting the semantic meaning and Bloom’s classification of learning outcome statements. The frequency with which words in learning outcomes appear as different parts of speech and at different Bloom’s levels is described as a preliminary step of a larger project that aims to automatically classify Bloom’s levels using a combination of table lookup and machine learning approaches. It is indicated that automated parts of speech classification can assist human learning and teaching designers to write clearer learning outcome statements. This is in addition to playing a role in automated Bloom’s Taxonomy classification, and identifying cases requiring review in conjunction with normal institutional curriculum management processes.