Verification of Job Authenticity using Prediction of Online Employment Scam Model (POESM)

In the recent times it is found to that there is a growing interest in Internet of Things (IoT) and its respective sophisticated cloud architectures. However, such development does not ensure the confidentiality, integrity due to its security vulnerabilities in its improvements. Therefore data breaching occurs rapidly in all sectors. Nowadays, several job-seekers are becoming the prime victim for such digital loophole fraudsters, when there is an advertisement for bulk requirement. In order to reduce frauds and scams that are designed to take advantage of people who seek jobs, a model to predict the genuineness of posting digital vacancies for job is real or fake. This paper presents the implementation of Prediction of Employment Scam Model (POESM), intending for the classification of fraudulent and non-fraudulent digital job posting advertisements. In our work, we used eight techniques such as Logistic Regression, Naive Bayes, Multiple Layer Perceptron, K-Nearest Neighbor, Decision Tree, Random Forests, Adaboost, Gradient Boosting classifiers. To analyze the dataset, supervised machine learning techniques is used to capture several essential information. Empirical results are found based on Accuracy, F1-Score, Cohen-Kappa Score, and MSE declares that Random Forest classifier is best suited for the prediction of online fraudulent requirements. The proposed POESM strategy can be a promising solution for identifying the genuineness of vacancies for job posted via internet.