A Comparative Study on Fake Job Post Prediction

Abstract: Because of advancements in current innovation and social correspondence, publicising new job openings has recently become an exceptionally common issue in today's world. As a result, everyone will be concerned about the fake job posting expectation task. As with other grouping endeavours, counterfeit work presenting forecast brings with it a slew of difficulties. This paper proposed using various information mining methods and characterization calculations, for example, KNN, innocent bayes classifier, multi-facet perceptron, and profound brain organisation, to forecast whether a task post is genuine or fake. We examined the Employment Scam Aegean Dataset (EMSCAD), which contains 18000 examples. As a classifier, the profound brain network excels at this characterization task. For this powerful brain network classifier, we used three thick layers. The prepared classifier predicts a deceptive work post with 98 percent order exactness (DNN). Record Terms - bogus work expectation, profound learning, information mining Terms - bogus work expectation, profound learning, information mining.