International Journal of Progressive Research in Engineering Management and Science
(Peer-Reviewed, Open Access, Fully Referred International Journal)
www.ijprems.com
editor@ijprems.com or Whatsapp at (+91-9098855509)


FAKE JOB DETECTION USING MACHINE LEARNING (KEY IJP************086)
Abstract
In recent years, the rise of online job portals and platforms has provided job seekers with a convenient and efficient means of finding employment opportunities. However, this digital transformation has also given rise to a growing concern: the proliferation of fake job postings. These deceptive listings can mislead and exploit job seekers, wasting their time and potentially exposing them to fraudulent activities. This work tries to address this issue by performing a comparison between Logistic Regression, Multi-layer Perceptron, Random Forest, and Decision Trees algorithms to determine which automated system can accurately distinguish between legitimate job advertisements and fake postings. To perform our experimentation process, the Employment Scam Aegean Dataset (EMSCAD) dataset was used to train and test our models. To improve further our results, feature engineering was applied to the data set to create new features from raw data. Our results demonstrated that the Multi-layer Perceptron and Logistic Regression can accurately classify fake job posts. These models were the two that had the best results according to the accuracy, precision, recall, and f1-score which were the metrics we used to evaluate each of them. This research provides significant value to job seekers, employers, and job portals alike. By accurately detecting and filtering out fake job postings, job seekers can avoid potential scams and focus their efforts on genuine employment opportunities. Employers benefit from improved reputation and more qualified applicants, while job portals can enhance their credibility and trustworthiness.
DOI Requested