Classification of Genuinity in Job Posting Using Machine Learning

Abstract

Abstract: To avoid fraudulent Job postings on the internet, we target to minimize the number of such frauds through the Machine Learning approach to predict the chances of a job being fake so that the candidate can stay alert and make informed decisions if required. The model will use NLP to analyze the sentiments and pattern in the job posting and TF-IDF vectorizer for feature extraction. In this model, we are going to use Synthetic Minority Oversampling Technique (SMOTE) to balance the data and for classification, we used Random Forest to predict output with high accuracy, even for the large dataset it runs efficiently, and it enhances the accuracy of the model and prevents the overfitting issue. The final model will take in any relevant job posting data and produce a result determining whether the job is real or fake. Keywords: Natural Language Processing (NLP), Term Frequency-Inverse Document Frequency (TF-IDF), Synthetic Minority Oversampling Technique (SMOTE), Random Forest.