Benchmarking AI-Driven Resume Screening: an Evaluation of Precision and Efficiency

Abstract

In contrast to conventional rule-based and earlier ATS-based systems, this study introduces a revolutionary AI-based resume screening method that increases candidate selection accuracy and efficiency. On a variety of performance criteria, we evaluated the suggested model's performance against that of other methods, such as deep learning models, earlier ATS systems, and conventional rule-based systems. The efficiency of the suggested model in finding pertinent candidates was demonstrated by its 85% accuracy, 78% recall, and 81% F1 score. With an average processing time of only 0.5 seconds per resume, it processed 1,200 resumes every hour, which is an amazing rate. With a true positive rate (TPR) of 88% and a false positive rate of 10%, the model also demonstrated a high accuracy of 90%, suggesting that candidate selection mistakes were limited. Older ATS-based systems scored 75% in precision and 70% in recall, but traditional rule-based systems had lower precision (70%) and recall (65%). Although they needed more processing power, deep learning models fared better than the suggested solution in terms of recall (85%) and accuracy (90%). Additionally, the suggested approach showed high keyword matching accuracy (95%) and ATS compatibility (92%)— providing a scalable alternative for extensive hiring. All things considered, this study demonstrates how well AI can automate and enhance the hiring process.

Ähnliche Arbeiten