Real and Fake Job Classification Using NLP and Machine Learning Techniques

Detecting fraudulent job listings is a critical challenge in the recruiting industry. This paper presents a machine learning model for classifying real or fake job postings. The study begins with Exploratory Data Analysis (EDA) to gain insights into the multi-class classification of different attributes and understand their relationships. Next, data preprocessing techniques, including natural language processing (NLP), are employed to prepare the datasets for training and testing. Various machine learning methods such as K-Nearest Neighbors (KNN), Support Vector Machine (SVM), AdaBoost, Random Forest, Naive Bayes, and Logistic Regression are applied to classify job postings as real or fake. Performance evaluation measures including accuracy, precision, recall, F1-score, selectivity, and specificity are computed to assess the classifier’s effectiveness. The results demonstrate that the proposed model achieves an impressive accuracy of 99.2% in classifying job information using the random forest classifier. This highlights the robustness and efficacy of the model in accurately identifying fraudulent job listings, providing a valuable tool for job seekers, employers, and recruitment platforms in maintaining the integrity of the job market.