Fake Job Posting Detection Using Machine Learning: TF-IDF, Logistic Regression, and Real- Time Risk Assessment
Suresh L,
Neetha M,
Rakesh M,
Pranith Reddy K,
Reshma O
This paper presents the design and development of a Fake Job Posting Detection System developed using
Python and the Streamlit framework to protect job seekers from fraudulent employment advertisements
across digital recruitment platforms. The system employs supervised machine learning techniques
combining TF-IDF vectorization and Logistic Regression classification to analyze job posting content
and classify it as genuine or fake in real time. Features such as URL-based job content scraping, text
preprocessing using NLTK, risk score generation, and an admin monitoring dashboard are integrated to
enhance usability and security. The system is designed with a modular architecture and session-based
data handling for efficient prediction management. It is observed that the proposed system improves
job seeker safety, reduces exposure to online employment scams, and enhances user awareness about
fraudulent job postings.
