Assessing Question Quality using Bloom's Taxonomy
Bloom’s Taxonomy has been in existence for over half a century. The main idea behind this model was to promote higher forms of thinking in education, such as analyzing and evaluating concepts, processes, procedures, and principles, rather than simple rote learning. In evaluating the usefulness of Bloom’s Taxonomy, only the cognitive domain is usually considered. This is a well-researched topic and much work has been done on the subject. Bloom’s Taxonomy has proved to be a good pedagogic tool in the field of education. However, in evaluating information, even the cognitive processes and knowledge level must be considered. This version of the taxonomy considering both the cognitive domain as well as the cognitive processes dimension is called the Bloom’s Modified Taxonomy. The aim of this project will be to explore the various techniques in which the modified taxonomic principles can be applied, and to analyse the accuracy of each of these applications.
● Improve the knowledge accuracy of n-squared with SVM or GaussianNB. Also compare with LSA and LDA (next)
● Voting System for Skill/Knowledge classifiers
● [Aspirational] Test on college Question Papers
GOAL: Get a model with 60% accuracy or higher, for both the dimensions (validated with human recall).