Resume duplicate checking method and device, equipment and medium

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a resume duplicate checking method and device, equipment and a medium. The method comprises: acquiring a to-be-duplicated resume; performing word segmentation according to the to-be-duplicated resume, and performing hash signature matrix calculation on a word segmentation result to obtain a to-be-duplicated hashsignature matrix; according to the to-be-duplicated hash signature matrix, and carrying out similar resume query from a resume library according to information classification to obtain a candidate resume set; respectively constructing a resume pair feature vector for the to-be-duplicated resume and each resume in the candidate resume set to obtain a plurality of to-be-predicted resume pair feature vectors; inputting the to-be-predicted resume pair feature vectors into the classification prediction model for similarity probability prediction to obtain probability prediction values of the to-be-predicted resume pair feature vectors; and determining a target repeated resume pair according to the probability prediction values of the plurality of to-be-predicted resume pair feature vectors. According to the invention, the duplicate checking efficiency is improved, similar rules do not need to be set manually, and the accuracy of determining the target repeated resume pairs is guaranteed.

Description

technical field [0001] This application relates to the technical field of artificial intelligence, in particular to a resume checking method, device, equipment and medium. Background technique [0002] In the enterprise recruitment scenario, in order to prevent the same candidate from submitting to different positions at the same time, resulting in multiple recruitment processes and wasting the recruitment cost of the enterprise, it is necessary to repeatedly judge the resumes delivered and the resumes in the process; at the same time, enterprises are purchasing resumes from multiple channels In the process, in order to prevent repeated purchases, it is also necessary to repeatedly judge the resumes that need to be purchased in the internal resume database of the enterprise. [0003] Compared with the general text, the resume text is more changeable. For example, the same person may use different templates for their resumes in different periods; for another example, when th...

AI Technical Summary

Problems solved by technology

[0005] The main purpose of this application is to provide a method, device, equipment, and medium for checking plagiarism of resumes, aiming to solve the technical problem that existing text checking methods for checking plagiarism of resumes are difficult to apply to resume checks due to the changeable resume text in the prior art

Images