Online Recruitment Fraud Detection Using Deep Learning Approaches

Abstract

Online recruitment platforms have become widely used for job searching and hiring processes. However, the increasing popularity of these platforms has also led to a rise in fraudulent job postings that mislead job seekers and cause financial and personal data losses. This paper presents a deep learning-based approach for detecting online recruitment fraud by analyzing job descriptions and related textual information. Advanced natural language processing techniques and deep learning models such as GPT-2, XLNet and Long Short-Term Memory (LSTM) are utilized to classify job postings as genuine or fraudulent. Experimental results show that the proposed system effectively improves detection accuracy and reduces the risk of online recruitment fraud. The proposed framework focuses on extracting meaningful linguistic and contextual features from job advertisement to identify hidden fraud patters that are difficult to detect using traditional machine learning methods. Text preprocessing, tokenization and embedding techniques are applied to convert unstructured job posting data into numerical representations suitable for deep learning models. By leveraging transformer-based architectures along with sequential neural networks, the system captures both semantic meaning and contextual dependencies present in recruitment content. This enables the model to differentiate subtle differences between legitimate and deceptive job descriptions.  Furthermore, the developed fraud detection model contributes to enhancing trust and safety in online recruitment environments by enabling automated screening of job postings at scale. The approach can assist recruitment platforms and regulatory authorities in identifying suspicious listings before they reach potential applicants. The experimental evaluation demonstrates that deep learning-based classification significantly outperforms conventional approaches in terms of precision and recall. The proposed system can be integrated into real-world recruitment platforms to provide early fraud detection, thereby protecting job seekers and improving the reliability of online hiring ecosystems.