Published Paper Details:

The surge in online job portals has greatly expanded access to employment opportunities worldwide. However, this increased convenience has also led to a rise in fraudulent job postings, putting job seekers at risk of identity theft, financial scams, and other threats. This paper introduces an integrated Fake Job Detection and Sentiment Analysis System aimed at improving the credibility of job listings through the application of machine learning and natural language processing techniques. The system utilizes a Random Forest Classifier, trained on the Fake Job Post dataset, achieving a detection accuracy of 98%. To capture user sentiment, a Bidirectional Long Short-Term Memory (Bi-LSTM) model is trained on the Glassdoor Review dataset, reaching a sentiment classification accuracy of 63%. The proposed dual-layered architecture supports real-time authenticity validation and sentiment-based feedback analysis, enhanced by an intuitive feedback interface and an administrative dashboard for manual review and trend tracking. Unlike traditional approaches that treat detection and sentiment analysis as separate components, our system unifies both into a cohesive, scalable platform. It is adaptable to diverse job markets and offers potential applications across job portals, recruitment sites, and employer branding initiatives, fostering greater trust and minimizing users' exposure to fraudulent employment opportunities.