Question and answer method and system based on feature classification

• Main Authors: XIE ZHIHAO, WU SHUHAO, WEI ZHONGRUI, ZHANG ZISHUO, GAO SHENGNAN, ZHANG SHICHAO, HU MINGZHE, PENG LONG, GE XIANGYU, SONG BIAO, LI SHENGCHANG, SHAO PENGZHI, LI ZEYU, MENG YINGQIAN, REN ZHIYING, JIANG WEI
• Format: Patent
• Language: Chinese; English
• Published: 23.06.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention relates to a question answering method and system based on feature classification, and belongs to the technical field of natural language processing. The problems that in the prior art, a question and answer model is low in inference speed, and answer prediction is not accurate enough are solved. The question answering method comprises the following steps of obtaining a question to be processed, and obtaining a category corresponding to the question to be processed; searching original text data with the same category as the to-be-processed problem and the highest relevancy in a text corpus; the text corpus comprises a plurality of different types of original text data; fusing the to-be-processed question and the corresponding original text data with the maximum relevancy with the corresponding category labels to obtain two corresponding feature vectors with category label information, inputting the two feature vectors into an answer prediction module of a pre-trained intelligent question and ans