Comprehensive Review Of Detection Of Fake Job Listing Using AI

Abstract

Digital hiring platforms have revolutionized how companies recruit and how candidates search for work, yet this shift has catalyzed a sharp increase in recruitment scams. Cybercriminals frequently deploy fraudulent job advertisements to harvest sensitive user data, orchestrate financial fraud, and exploit unsuspecting applicants. In response, recent scholarship has focused heavily on automating the detection of these deceptive postings by leveraging Artificial Intelligence (AI), Machine Learning (ML), and Natural Language Processing (NLP). This literature review synthesizes current research evaluating various machine learning and deep learning methodologies designed to spot fake listings. Specifically, we examine the implementation of foundational algorithms including Logistic Regression, Support Vector Machines (SVM), Random Forest, and Naïve Bayes alongside more advanced architectures like XGBoost and Bidirectional Long Short-Term Memory (Bi-LSTM) networks. Past studies underscore that meticulous text preprocessing, metadata evaluation, and robust feature extraction (such as TF-IDF) are pivotal to maximizing classifier performance. Furthermore, the consensus among researchers points to ensemble methods and deep learning structures as the most accurate tools for capturing subtle, deceptive recruitment indicators. However, the literature also reveals persistent bottlenecks: severe dataset class imbalances, rapidly adapting fraudulent strategies, and the vulnerabilities inherent in analyzing text in isolation. Ultimately, this analysis confirms that AI-driven frameworks offer a scalable, dependable defense against hiring scams, reinforcing the need for intelligent verification systems to restore institutional trust and user safety in digital employment ecosystems.