Scholarlink

As of December 1, 2024, there are around 5333 employment organizations in India, and according to one research study, over 16,000 applications were received every day for 20 technical positions, averaging 800 applications per job. Another poll found a 48% increase in applications per job when compared to pre-pandemic levels, indicating a considerable rise in employment application volume. The main objective of this paper is to propose and develop a resume screening framework model especially built for IT job role prediction, which may assist job recruiters in hiring individuals for the relevant job role and job searchers in obtaining the suitable job. Furthermore, our resume screening dataset includes 48 distinct IT job positions that are relevant to the current IT employment market. We have proposed our Resume Screening Framework model and implemented it with the help of machine learning and NLP techniques. On the dataset, several ML-supervised and ensembling classifiers are used. Parameters such as Area Under Curve (AUC), Recall, Cohen Kappa score, F1 score, accuracy, and precision are used to gauge the performance of algorithms for the resume screening dataset. With an AUC value of 0.999 and an accuracy of 0.9792, the comparative findings unequivocally demonstrate that the CatBoost and LightGBM algorithms perform exceptionally well in the Resume Screening Dataset. Here, the ROC graphs for both datasets are demonstrated as well. We have also compared our resume screening model with other resume screening systems. The comparison shows that with an accuracy of 97.92%, our resume screening model is very accurate and precise to perform the IT job role prediction on any user’s resume.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Using Machine Learning to Automate IT Job Role Prediction by Resume Screening

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Prerna Agrawal,
  • Savita Gandhi,
  • Vidhi Dattani,
  • Darshan Rachchh,
  • Mahek Rathod,
  • Khushi Vadher

摘要

As of December 1, 2024, there are around 5333 employment organizations in India, and according to one research study, over 16,000 applications were received every day for 20 technical positions, averaging 800 applications per job. Another poll found a 48% increase in applications per job when compared to pre-pandemic levels, indicating a considerable rise in employment application volume. The main objective of this paper is to propose and develop a resume screening framework model especially built for IT job role prediction, which may assist job recruiters in hiring individuals for the relevant job role and job searchers in obtaining the suitable job. Furthermore, our resume screening dataset includes 48 distinct IT job positions that are relevant to the current IT employment market. We have proposed our Resume Screening Framework model and implemented it with the help of machine learning and NLP techniques. On the dataset, several ML-supervised and ensembling classifiers are used. Parameters such as Area Under Curve (AUC), Recall, Cohen Kappa score, F1 score, accuracy, and precision are used to gauge the performance of algorithms for the resume screening dataset. With an AUC value of 0.999 and an accuracy of 0.9792, the comparative findings unequivocally demonstrate that the CatBoost and LightGBM algorithms perform exceptionally well in the Resume Screening Dataset. Here, the ROC graphs for both datasets are demonstrated as well. We have also compared our resume screening model with other resume screening systems. The comparison shows that with an accuracy of 97.92%, our resume screening model is very accurate and precise to perform the IT job role prediction on any user’s resume.