Published Paper Details:

In today's digital era, online recruitment platforms have made job searching easier, but they also present challenges such as irrelevant job recommendations and the increasing presence of fraudulent job postings. To address these issues, this project proposes a Resume-Centric Job Recommendation and Fake Job Detection System using machine learning techniques. The system analyzes uploaded resumes by extracting text from PDF files and identifying key skills and competencies through keyword processing. It then matches these skills with job descriptions obtained from real-time job search APIs. To measure the relevance between a candidate's profile and job descriptions, TF-IDF (Term Frequency-Inverse Document Frequency) is used for feature extraction, and cosine similarity is applied to compute a similarity score, enabling accurate and personalized job recommendations. Additionally, the system enhances user safety by detecting fraudulent job postings using a Logistic Regression classifier trained on labeled job data to identify suspicious patterns and classify listings as genuine or fake. The complete system is implemented using the Flask framework, providing an interactive and user-friendly web interface where users can upload resumes and filter job results based on criteria such as location and role. Experimental results demonstrate that the proposed system improves recommendation accuracy while significantly increasing trust and reliability in online recruitment platforms by effectively identifying fake job listings.