work history information extraction method based on a double-layer BiLSTM-CRF

• Main Authors: GAO JING, YIN JIAN, LIN CHUANGWEI, LAI HANJIANG
• Format: Patent
• Language: Chinese; English
• Published: 19.04.2019
• 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 work history information extraction method based on a double-layer BiLSTM-CRF which comprises the following steps: S1, preprocessing work history information; S2, according to the time division work history information as work experience, preprocessing the work experience; S3: utilizing double-layer BiLSTM-CRF model extracts the information entity of the work experience;S4, further processing the information entity extracted in the step S3; S5, arranging information. A Double-layer BiLSTM model is used in the invention so that the CRF model can better extract the information entity in the work experience. The problem that information extraction is difficult due to factors such as information entity intersection and Chinese information entity irregularity is better solved. Besides, a traditional information extraction task is divided into a plurality of sub-tasks, a disambiguation module and an association module are added, high aggregation and low coupling are achieved, extraction ca