Natural language processing method and device applied to invoice information mining

• Main Author: HUANG ZEYUN
• Format: Patent
• Language: Chinese; English
• Published: 07.06.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a natural language processing method and device applied to invoice information mining, relates to the technical field of natural language processing, and aims to solve the problem that an existing natural language processing technology is not suitable for invoice information extraction. The method comprises the steps of performing data cleaning on invoicing details of enterprise sales invoices, and extracting effective sales information; converting the cleaned invoicing details into high-dimensional word vectors; performing unsupervised clustering on the high-dimensional word vectors; performing manual integration and labeling on the unsupervised clustering result to obtain a learning sample; carrying out transfer learning on the natural language processing model by utilizing a learning sample, and training a model capable of identifying commodity type labels; inputting the enterprise sales invoice into the model to obtain a corresponding prediction result, and obtaining a sales profit