Random Forest based Fake Job Detection

The intersection of computer science and social sciences leads an increasing problem: fraudulent job ads. Their presence can have far-reaching negative ramifications that must be dealt with immediately in order to limit further harm to society. Due to the vast amounts of data accumulated worldwide, effective ways must be found for distinguishing legitimate job advertisements from fraudulent ones. This paper proposes a machine learning approach designed to differentiate between fraudulent and non-fraudulent job postings, offering an automated tool to mitigate the proliferation of fraudulent job listings on the internet. The methodology relies on a range of machine learning classification techniques to analyze web-based job postings, comparing the efficacy of different classifiers in identifying employment scams. The primary objective is to establish a robust model for the detection of fake job posts amidst the overwhelming volume of online listings. This research recommends employing various data mining and categorizing algorithms such as Decision Tree and Support Vector Machine, Naive Bayes classifier, Random-forest classifier and Multilayer Perceptron to detect whether an advertisement is genuine. This study has utilized Kaggle data with the proposed classifier enhancing 99.48% accuracy when classifying fake job ads. Additionally, this method provides techniques to detect employment scams to meet applicants' need to protect themselves against scammers.