Online Recruitment Fraud (ORF) Detection Using Deep Learning Approaches

Mrs. Mynam Sushma Devi, Nagaram Akshitha Goud, Mr.Vijay Prasad Tiwari, Chawan Rahul, Sathugari Shravya Reddy

doi:10.64751/

Authors

Mrs. Mynam Sushma Devi, Nagaram Akshitha Goud, Mr.Vijay Prasad Tiwari, Chawan Rahul, Sathugari Shravya Reddy Author

DOI:

https://doi.org/10.64751/

Keywords:

Class imbalance, data augmentation, deep learning, employment scam, fraud detection, machine learning, online recruitment, SMOTE, transformer-based models.

Abstract

Online recruiting platforms have transformed the employment process; yet, they have also led to the
emergence of bogus job listings, resulting in financial losses for job searchers. A deep learning algorithm is
suggested to identify online recruiting fraud (ORF) with a fresh dataset derived from Fake Job Posting, Pakistan
Job Posting, and US Job Posting datasets. The method utilizes Bidirectional Encoder Representations from
Transformers (BERT) and Robustly Optimized BERT Pre-training Approach (RoBERTa) to convert task
specifications into numerical vectors. To address the significant class imbalance in the dataset, the SMOTE
variation, SMOBD, is utilized for efficient class balancing. The experimental framework incorporates these
advanced characteristics with a two-dimensional Convolutional Neural Network (CNN2D) for job categorization.
The results indicate that the integration of BERT features and SMOBD with CNN2D attains the greatest
classification accuracy of 98.68%. This technique mitigates the shortcomings of obsolete datasets, offering a
comprehensive solution for identifying fake job listings and substantially aiding in the avoidance of online
recruiting frauds.