Engineering project word document-oriented intelligent information extraction method and system

• Main Authors: CHEN YANG, ZHANG LIBIN, BU XIANZHENG, CHANG XIUFENG, LI JIN, TAN PATRICK, CHEN YUANQI, LI YI, DING HAIBIN
• Format: Patent
• Language: Chinese; English
• Published: 12.09.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides an engineering project word document-oriented intelligent information extraction method and system, and the method comprises the steps: obtaining an engineering document, and carrying out the word segmentation processing to obtain document vocabularies; encoding the document vocabularies to obtain vocabulary vectors, and inputting the vocabulary vectors into a preset part-of-speech classification model to obtain a part-of-speech vector corresponding to each vocabulary vector; combining the part-of-speech vector and the vocabulary vector to obtain a combined vector corresponding to each vocabulary; the combined vectors are input into a preset item classification model, the item classification model outputs items corresponding to the combined vectors, and the items are all items in a preset table; and filling the document vocabulary corresponding to each combined vector into the preset table, and outputting the filled table. According to the scheme, the item corresponding to each combined