A Hybrid Approach to Fake Job Detection in Online Recruitment Platform

Recent advancements in online recruitment technology have allowed job seekers to utilize these services to research jobs more easily and quickly. While the ease of using these websites has increased for recruiters, it has also allowed for a rise in scam job postings for job seekers. These fake job postings can seem legitimate and use suspicious language, including promises for high pay, unofficial ways to communicate, and promises of quick employment; some of these posts can lead to theft of money or job seekers' private information. To help deal with this issue, we have created the JobShield, a hybrid job posting fraud detection system. JobShield is a hybrid system that utilizes machine learning algorithms along with rules-based verification to look for suspicious trends in job postings through job descriptions, recruiter info, and other fraud indicators to determine if a job posting is legitimate or not. The JobShield will also alert users when a job posting may be suspect and potentially not safe to apply to. JobShield was designed and built with a user friendly interface and is capable of running verifications in real time. Results from experimental trials indicate that the JobShield has an accuracy rate of approximately 85-90%, demonstrating the effectiveness of the system for identifying fraudulent job postings. This study establishes the need to use multiple detection techniques to improve the reliability and trustworthiness of online-recruitment platforms.