ABSTRACT
The use of machine learning models holds promise for expediting the hiring process in the dynamic sector of talent acquisition. Using a sizable dataset of 500 resumes and interview replies from 200 candidates for training and an additional 150 resumes and 70 candidate comments for model validation, this study explores the challenging areas of resume shortlisting and applicant selection. We thoroughly evaluate the performance of machine learning models in terms of accuracy, precision, recall, and F1-Score using Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), Naive Bayes (NB), Support Vector Machines (SVM), Random Forest (RF), and Decision Trees (DT). The results highlight the ANN's better performance, which includes a 97.5% accuracy rate, while showcasing the dynamic advantages of each model. A careful examination of confusion matrices also demonstrates the trade-offs each model makes between recall and precision. For businesses looking to improve their hiring practices through data-driven decision-making, these findings present new opportunities. This study stresses how machine learning can streamline the hiring process and stay up with talent acquisition requirements in today's competitive employment market.
