Comprehensive Approach to Fraudulent Job Post Detection Using Machine Learning and BERT Models

In recent times there has been a significant rise in fraudulent job postings posing risks to job seekers and recruitment platforms. The paper presents a framework combining Machine Learning (ML) and advanced Natural Language Processing (NLP) techniques to detect the fake job postings. The model uses data that has been integrated from multiple sources and uses traditional ML models and NLP models. Feature extraction methods such as fraud keyword detection and sentiment polarity analysis were done using TextBlob and VADER to enhance interpretability. Results show that the BERT model has achieved 100% accuracy by outperforming the other models and also shows the system’s ability to differentiate between real and fake job postings. Comparative analysis was done to show the superiority of deep learning models over the traditional approaches. The system provides a robust solution to enhance job security, prevent cyber fraud, and assist recruitment platforms in identifying fraudulent job postings.