Question Classification Using Profile Hidden Markov Models

Abstract

Recently, Question Answering has been a hot topic in the research of information retrieval. Question Classification plays a critical role in most Question Answering systems. In this paper, a new approach to classifying questions using Profile Hidden Markov Models (PHMMs) is proposed. The generalization strategies to extract the pattern instances of questions by selective substitution are discussed. Then the classification method with pattern instances' structural features is investigated. Experimental results show that the PHMM based question classifier can reach the accuracy of 92.2% and significantly outperforms most of the state-of-the-art systems.

Feature extraction based on information gain and sequential pattern for English question classification

“…Many well‐known classification approaches, including k ‐nearest neighbours, naive Bayes’ (NB), decision trees (DTs), maximum entropy (ME), support vector machines (SVMs), hidden Markov model [33], and kernel [34] have been employed for QC. It has been proved through experimentation that the performances of ME and SVM are superior to k ‐NN, NB, and DT.…”

Garbage FusionNet: A deep learning framework combining ResNet and Vision Transformer for waste classification

“…Mimicking the neural network architecture of the human brain, deep learning algorithms are adept at autonomously extracting features and discerning patterns from extensive datasets. This is particularly evident in the realm of image classification, where deep learning approaches have shown superior efficacy, adeptly pinpointing and categorizing objects amidst intricate backgrounds [12][13][14][15] . When juxtaposed with conventional manual methods of waste classification, systems predicated on deep learning offer manifold benefits.…”