http://hdl.handle.net/123456789/9927 Fake Job Prediction 2026-04-16T18:13:50Z http://hdl.handle.net/123456789/9928 Fake Job Prediction Mathew, Merin Now a days it becomes extremely difficult for one to stay safe online. Number of victims of fake job posting is increasing drastically day by day. Identify the fraudulent job advertisement manually is very difficult. To avoid fraudulent post for job in the internet, an automated tool using machine learning based classification techniques is proposed in the project. This study research attempts to prohibit privacy and loss of money for individuals and organization by creating a reliable model which can detect the fraud exposure in the online recruitment environments. This research presents a major contribution represented in a reliable detection model using ensemble approach based on Random forest classifier to detect Online Recruitment Fraud (ORF). ORF detection is an important problem to solve but it has not received much attention from the research community and it is currently a relatively unexplored area. Detection of fraud job offers from a legitimate set of job is a technically challenging problem. The main challenge the class imbalance problem as the number of fraud jobs are relatively less as compared to the legitimate jobs. This makes learning the features of fraud jobs for automated prediction a challenging task. These fraudulent job post detection draws a good attention for obtaining an automated tool for identifying fake jobs and reporting them to people for avoiding application for such jobs. Different classifiers are used for checking fraudulent post in the web and the results of those classifiers are compared for identifying the best employment scam detection model. It helps in detecting fake job posts from an enormous number of posts. Random Forest classifier can be regarded as the best model for this fake job detection scheme. Random forests creates decision trees on randomly selected data samples, gets prediction from each tree and selects the best solution by means of voting. It also provides a pretty good indicator of the feature importance. The proposed model achieved an accuracy 95%. 2022-06-11T00:00:00Z