Short-Text Semantic Similarity Model of BERT-Based Siamese Network

People convey their emotions and thoughts through words, the medium of human thoughts. Up against the vigorous development of streaming media, the calculation of text similarity is imperative in the field of natural language processing. Any text-related field is inseparable from text semantic similarity. The calculation of text semantic similarity plays a key role in document management, document classification, and document relevance. Besides, popular natural language processing tasks in some trendy fields, such as artificial intelligence, human-machine translation, problem system, intelligent chat system, and nomenclature recognition, are intertwined with text semantic similarity calculation. In recent years, many excellent researchers have studied the algorithms and models of text semantic similarity from different dimensions. In this paper, a new short-text cosine similarity calculation model of the BERT-based Siamese network is proposed.