Resume Parsing based on Multi-label Classification using Neural Network models

Document Type

Conference Proceeding

Publication Date

5-22-2021

Abstract

Application for jobs usually brings much work for both appliers and HR. Appliers want to apply for the jobs which they are most suitable. The number of applications for a particular position can be significant, making the candidates' selection cumbersome for HR. Nowadays, hiring processes are often conducted through the Virtual mode with emails. This creates chances for analyzing the data in the resume. Therefore, to enhance selection problems' efficiency, resume parsing algorithms have been developed in recent years to predict resume-based skills or good jobs quickly. The artificial neural network is a hot spot in the field of artificial intelligence since the 1980s. It abstracts the human brain's neural network from the angle of information processing, establishes some simple models, and forms different networks according to different connection modes. In recent years, neural networks-based algorithms perform high efficiency in processing text classification. This paper put forward some of the efficient algorithms used in text classification, Like BPNN, CNN, BiLSTM, and CRNN, for resume parsing. The original resumes are parsed by splitting them into words, and word base is trained to get the most appropriate word, which has a high score in the resume is resulting suitable job for each resume. The CRNN performs best in resume parsing, which the accuracy can reach 96%. CNN places the lowest accuracy. The BPNN achieves good accuracy but brings inflexible.

Publication Title

ACM International Conference Proceeding Series

First Page Number

177

Last Page Number

185

DOI

10.1145/3469968.3469998

This document is currently not available here.

Share

COinS