AI-Powered Detection of Fraudulent Job Listings using DistilBERT and XGBoost

Fake job ads pose significant danger to job hunters on the Internet because they can steal personal data or money. This research demonstrate a model which is AI-driven using DistilBERT for text understanding and XGBoost for classification which successfully identifies the fraudulent job postings. NLP is employed to understand the job descriptions in addition to the characteristics such as the type of telecommuting job. The application of SMOTE(Synthetic Minority Oversampling Technique) to the data in combination with the advanced machine learning methods led to the successful model outcome, producing high accuracy and a reliable outcome. The research concludes that the system, which is able to identify fake job advertisements, thus reducing the number of fraud occurrences on the job market, is a feasible solution to online job market safety.