The project proposes an application that uses machine learning-based categorization approaches to avoid bogus job postings on the internet. Different classifiers are employed to check for fraudulent posts on the web, and their findings are compared to determine the best employment scam detection model. It aids in the detection of bogus job postings among a large number of postings. The article examines the various strategies used to tackle the bogus job posting on the internet. A survey for each and every approach selected for locating this fake job positing communications from the internet, and finally, we attempt to identify the problem gap between each and every works that has already been published on this topic. We collected several well-known papers related to our topic from 2004 to 2022, taking into account the similarities between each and every approach for effective detection, and finally attempting to determine which mechanism is best in providing fake job detection. For the identification of fake job postings, two basic types of classifiers are considered: single classifiers and ensemble classifiers. However, experimental results show that ensemble classifiers outperform single classifiers in detecting scams.