Improving Imbalanced Question Classification Using Structured Smote Based Approach

Questions Classification (QC) is one of the most popular text classification applications. QC plays an important role in question-answering systems. However, as in many real-world classification problems, QC may suffer from the problem of class imbalance. The classification of imbalanced data has been a key problem in machine learning and data mining. In this paper, we propose a framework that deals with the class imbalance using a hierarchical SMOTE algorithm for balancing different types of questions. The proposed framework is grammar-based, which involves using the grammatical pattern for each question and using machine learning algorithms to classify them. Experimental results imply that the proposed framework demonstrates a good level of accuracy in identifying different question types and handling class imbalance.