Text classification method and system based on cue words

• Main Authors: ZHOU ZHENGJIE, ZHAO YI, FENG XIAOMIN, QU YI, SUN KE, LI YIFEI, CUI YUAN, YU YINTAO, WEI ZHILI
• Format: Patent
• Language: Chinese; English
• Published: 16.05.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides a text classification method and system based on cue words, and belongs to the technical field of computer technology and text classification. The method comprises the following steps: generating a cue word according to a task attribute, and splicing the cue word with a first text to obtain a second text; word segmentation is carried out on the second text to obtain vector data, and the vector data comprises word representation vectors and position representation vectors; transmitting the vector data to a subject extraction model for encoding, and performing binary classification on each prediction flag bit of the cue word to obtain a plurality of classification results; and analyzing and integrating all the classification results to obtain a classification result of the first text. According to the scheme, continuous learning of the text multi-classification model is realized, and the universality of the model is improved. 本发明提供了一种基于提示词的文本分类方法及系统，属于计算机技术及文本分类技术领域；其中，所述方法包括：根据任务属性生成提示词，