A survey of text classification based on pre-trained language model

The utilization of text classification is widespread within the domain of Natural Language Processing (NLP). In recent years, pre-trained language models (PLMs) based on the Transformer architecture have made significant strides across various artificial intelligence tasks. Currently, text classification employing PLMs has emerged as a prominent research focus within NLP. While several review papers examine text classification and Transformer models, there is a notable lack of comprehensive surveys specifically addressing text classification grounded in PLMs. To address this gap, the present survey provides an extensive overview of text classification techniques that leverage PLMs. The primary components of this review include: (1) an introduction, (2) a systematic examination of PLMs, (3) deep learning-based text classification methodologies, (4) text classification approaches utilizing pre-trained models, (5) commonly used datasets and evaluation metrics in text classification, (6) prevalent challenges and emerging trends in the field, and (7) a conclusion.