ABSTRACT
COVID-19 pandemic is behind the implementation of the "AI recruitment system.” The number of companies trying to introduce AI recruitment systems is increasing because the non-face-to-face method is recommended due to the COVID-19 pandemic and the management change of the organization comes with the development of IT technology. Behind the positive evaluation that the development of AI technology improves the efficiency of work, the demand for fair and transparent recruitment procedures has been increasing as controversy over fairness and objectivity has increased due to various hiring irregularities. This study aimed to approach in a more systematic and scientific way to maximize the effect of recruiting talent. In the previous study, voice and video were identified based on ML. In situations where the problem of truth and falsehood is raised, this study conducted EEG-based biological experimental studies with a deep learning method to explore more objectively. Also, the experimental design applied biological experiments between brain activity patterns and brain regions as signals from EEG-based 14 channels to explore the truth/false authenticity of the experimenters. As a result of the experiment, the best performance and effect were shown in the CNN model with an accuracy of 91% truth and 89% false among the comparative analysis of Decision Tree, Random Forest, and CNN. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
ABSTRACT
COVID-19 pandemic is behind the implementation of the “AI recruitment system.” The number of companies trying to introduce AI recruitment systems is increasing because the non-face-to-face method is recommended due to the COVID-19 pandemic and the management change of the organization comes with the development of IT technology. Behind the positive evaluation that the development of AI technology improves the efficiency of work, the demand for fair and transparent recruitment procedures has been increasing as controversy over fairness and objectivity has increased due to various hiring irregularities. This study aimed to approach in a more systematic and scientific way to maximize the effect of recruiting talent. In the previous study, voice and video were identified based on ML. In situations where the problem of truth and falsehood is raised, this study conducted EEG-based biological experimental studies with a deep learning method to explore more objectively. Also, the experimental design applied biological experiments between brain activity patterns and brain regions as signals from EEG-based 14 channels to explore the truth/false authenticity of the experimenters. As a result of the experiment, the best performance and effect were shown in the CNN model with an accuracy of 91% truth and 89% false among the comparative analysis of Decision Tree, Random Forest, and CNN. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.