FAKE JOB DETECTION SYSTEM USING MACHINE LEARNING

Abstract

Abstract -The proliferation of online job recruitment platforms has brought unprecedented convenience to job seekers and employers. However, this convenience comes with a downside - the rise of fake job postings, which not only waste job seekers' time but also pose financial risks. To address this challenge, our project focuses on developing a fake job prediction system using machine learning techniques. Unlike previous versions, our system utilizes Streamlit, a Python library for building user interfaces, as the primary framework for the frontend. Users interact with the application directly through Streamlit, providing job postings for analysis. The core of our system lies in a trained machine learning model, which is saved as a pickle file. This model processes the user-inputted job postings and provides predictions on their legitimacy. The project's workflow involves user input, processing through the trained model, and outputting the predictions back to the user interface. Through meticulous analysis of various features extracted from job postings, our model aims to deliver accurate predictions in real-time, thereby empowering job seekers to make informed decisions and safeguarding the credibility of online recruitment platforms. This abstract provides an overview of our project's objectives, methodology, and outcomes. Through the integration of Streamlit and a trained machine learning model, we aim to provide a user- friendly and efficient solution for detecting fake job postings, ultimately contributing to a safer and more trustworthy online job market.