A Comparative Study on Fake Job Post Prediction Using Different Data Mining Techniques

Abstract

With the development of social media and contemporary technologies, advertising
new job openings has recently become a very prevalent problem in the current world.
Therefore, everyone will have a lot of reason to be concerned about bogus job postings. Fake
job posing prediction presents a variety of difficulties, much as many other categorization
problems. In order to determine if a job posting is legitimate or fake, this study proposes
using several data mining methods and classification algorithms such KNN, decision tree,
support vector machine, naive bayes classifier, random forest classifier, multilayer
perceptron, and deep neural network. 18000 samples from the Employment Scam Aegean
Dataset (EMSCAD) were used in our experiments. For this classification challenge, a deep
neural network classifier excels. For this deep neural network classifier, three thick layers
were employed. A bogus job advertisement may be predicted with a classification accuracy
of around 98% by the trained classifier using DNN.