Interview entity recognition model training method and device and interview information entity extraction method and device

Abstract

The invention relates to the technical field of prediction models, and discloses an interview entity identification model training method and device and an interview information entity extraction method and device. According to the method, standard label prediction is carried out on first interview sample data through a direct prediction module in a preset identification model; standard label distribution and interview coding vectors are obtained; auxiliary label prediction is carried out on the first interview sample data according to the coding vector through each auxiliary prediction module to obtain auxiliary label distribution output by each auxiliary prediction module; a total loss value of a preset identification model is determined according to the auxiliary label distribution and the standard label distribution; and when the total loss value does not reach the preset convergence condition, a first initial parameter of the iterative preset identification model is updated until the total loss value reaches the preset convergence condition, and the converged preset identification model is recorded as an interview entity identification model. According to the invention, the model training efficiency and the model identification accuracy are improved.

Description

technical field [0001] The present invention relates to the technical field of prediction models, in particular to a method, device, equipment and medium for training interview entity recognition models and extracting interview information entities. Background technique [0002] The essence of named entity recognition is a sequence labeling problem. It is to input a sentence and output the entity corresponding to each word in the sentence, that is, to identify entities with specific meaning in the document, such as person names, place names, school names and proper nouns. For example, for the self-introduction of the interviewer in the intelligent recruitment process, it may be necessary to extract the company name and school name, so as to facilitate subsequent extraction and use of the interviewer's information. [0003] At present, for the problem of named entity recognition, if named entity recognition is performed through supervised learning, the demand for data volume ...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides an interview entity recognition model training, interview information entity extraction method, device, equipment and medium to solve the problem that supervised learning has a very large demand for data and unsupervised learning limits the accuracy of the model

Images