A Deep Learning Framework for Detecting Fraudulent Online Job Postings

Abstract

Online recruitment platforms have become a primary medium for job searches, but the increasing volume of fraudulent job postings poses serious risks to job seekers, including identity theft, financial loss, and data exploitation. Traditional rule-based and machine learning methods often fail to detect sophisticated or newly emerging fraud patterns due to the dynamic nature of online scams. This study presents a Deep Learning Framework for Detecting Fraudulent Online Job Postings, designed to automatically learn complex linguistic, semantic, and behavioral patterns associated with recruitment fraud. The proposed system integrates advanced neural architectures—including Bidirectional LSTM, CNN-based text feature extractors, and attention mechanisms—to capture subtle anomalies in job descriptions, employer metadata, and posting behaviors. A large, preprocessed dataset of legitimate and fraudulent job postings is used to train and evaluate the framework. Experimental results demonstrate that the deep learning–based model significantly outperforms traditional machine learning baselines in terms of accuracy, precision, recall, and F1-score, achieving robust detection of deceptive content. This framework contributes an intelligent, scalable, and automated solution, enhancing the safety and trustworthiness of online recruitment platforms while reducing the risk of applicant exploitation.