Comparative Analysis of Deep Learning and Machine Learning Techniques for Question Classification in Bloom's Taxonomy

In the current digital era, automation of question classification based on cognitive level has become important, especially with the increasing adoption of digital learning methods. This study investigated and compared the capabilities of various Deep Learning (DL) and Machine Learning (ML) techniques in question classification according to Bloom's Taxonomy. The four models considered in this analysis include Long Short-Term Memory (LSTM), Recurrent Neural Networks (RNN), Random Forest (RF), and K-Nearest Neighbor (KNN). The research methodology involves developing and training each model with a data set of questions pre-categorized based on Bloom's Taxonomy levels. After the training phase, each model is evaluated based on its accuracy in classifying questions into appropriate categories. The results show that the LSTM model provides the best performance with an accuracy of 83%, followed by RNN with 74%. Meanwhile, traditional machine learning techniques, RF and KNN, achieve 68% and 35% accuracy.