Self-supervised and Few-shot Contrastive Learning Frameworks for Text Clustering

Contrastive learning is a promising approach to unsupervised learning, as it inherits the advantages of well-studied deep models without dedicated and complex model design. In this paper, based on bidirectional encoder representations from transformers (BERT) long-short term memory (LSTM) neural networks, we propose self-supervised contrastive (SCL) well few-shot (FCL) with data augmentation (UDA) for text clustering. BERT-SCL outperforms state-of-the-art clustering approaches short texts long in terms several evaluation measures. LSTM-SCL also shows good performance BERT-FCL achieves close supervised UDA further improves texts. LSTM-FCL Our experiment results suggest that both SCL FCL are effective