AN INTELLIGENT DEEP NEURAL NETWORK APPROACH FOR FRAUDULENT JOB POSTING DETECTION

Abstract

This investigation employs Deep Neural Networks (DNNs) to identify fraudulent job postings on recruitment portals. The rise in fraudulent job advertisements is attributable to the expansion of online job boards. The objective of these advertisements is to deceive individuals into falling victim to financial scams, identity theft, and fraud. The proposed method is capable of detecting deceptive advertisements by utilizing linguistic and behavioral characteristics that are derived from corporate profiles, job postings, salary information, and posting patterns. The concealed fraud patterns are automatically identified by DNN models that have been trained on extensive recruitment datasets. Consequently, it is capable of accurately identifying job postings that are fraudulent. Feature embedding, data standardization, and tokenization are examples of advanced input techniques that improve model performance and reduce the number of incorrect predictions. The results indicate that the deep neural network surpasses the market's conventional machine learning techniques in terms of object detection, memory, and accuracy. An scalable and dependable solution that enhances the security, transparency, and dependability of online employment platforms safeguards job seekers from illegal recruitment and cyberfraud. Furthermore, this solution safeguards applicants from being exploited.