Detecting Fraud Job Recruitment Using Features Reflecting from Real-world Knowledge of Fraud

A common method for text-analysis and text-based classification is to process term-frequency or patterns of terms. However, these features alone may not be able differentiate fake authentic job advertisements. Thus, in this work, we proposed a detect recruitments using novel set designed reflect the behavior fraudsters who present information. The were missing information, exaggeration, credibility. represent form category an automatically generatable score readability. Data from EMSCAD dataset transformed accordance with used train detection model detection. experimental results showed that performed better than those based on approach every applied machine learning technique. yielded 97.64% accuracy, 0.97 precision 0.99 recall its best when classifying