A smart secured framework for detecting and averting online recruitment fraud using ensemble machine learning...

Area of Science:

• Computer Science
• Cybersecurity
• Data Science

Background:

• The internet and social media facilitate information access but also enable online recruitment fraud (ORF).
• ORF victims globally lose privacy due to fraudsters collecting personal data via fake job postings.
• Automated systems are needed to detect and prevent ORF, safeguarding job seekers.

Purpose of the Study:

• To develop a smart, secure framework for detecting and preventing online recruitment fraud.
• To leverage ensemble machine learning techniques for identifying fraudulent job advertisements.
• To enhance applicant privacy by distinguishing legitimate from fake job postings.

Main Methods:

• Utilized four ensemble machine learning methods: AdaBoost (AB), Xtreme Gradient Boost (XGB), Voting, and Random Forest (RF).
• Pre-processed a dataset using cleaning and denoising techniques to optimize detection performance.
• Evaluated model performance using accuracy, precision, sensitivity, F-measure, and ROC curves.

Main Results:

• AdaBoost (AB) demonstrated superior performance among the tested ensemble methods.
• The proposed AB framework achieved a high accuracy of 98.374% in detecting online recruitment fraud.
• Comparative analysis with existing methods confirmed the reliability and effectiveness of the AB model.

Conclusions:

• Ensemble machine learning, particularly AdaBoost, offers a reliable solution for detecting and preventing online recruitment fraud.
• The developed framework effectively distinguishes fake job advertisements, thereby protecting job applicants' privacy.
• This research contributes a robust model for enhancing cybersecurity in online recruitment processes.