A review questions classification based on Bloom taxonomy using a data mining approach

Abstract

Bloom's taxonomy is used to categorize learning objectives into various cognitive levels. This study discusses the role of ontology in the classification of Bloom's taxonomy-based questions using a computer science approach in text mining. This research aims to review and analyze using a systematic ontology approach in cognitive level question classification techniques using Bloom's taxonomy with a text-mining scientific approach. Based on the prism method, 22 papers were analyzed from 490 articles from databases such as Scopus, ACM, IEEE, Springer, and Elsevier, published in 2016-2023. Meanwhile, qualified experts have not validated the main factors influencing the application of taxonomy-based question classification. Based on the evaluation results of using traditional, deep learning, and hybrid models in single-class question classification, it provides higher accuracy than in multiple classes in the case of bloom taxonomy. In various classification models, there is no significant difference in accuracy in the algorithm; the difference in results occurs due to data imbalance problems in multiple classes in the case of bloom taxonomy. This case provides a considerable opportunity to explore the topic of Bloom's taxonomy in the knowledge discovery database in KDD databases