Back to Search Start Over

Identifying fake job posting using selective features and resampling techniques.

Authors :
Afzal, Hina
Rustam, Furqan
Aljedaani, Wajdi
Siddique, Muhammad Abubakar
Ullah, Saleem
Ashraf, Imran
Source :
Multimedia Tools & Applications; Feb2024, Vol. 83 Issue 6, p15591-15615, 25p
Publication Year :
2024

Abstract

The fake job posting has emerged as an alarming cyber-crime during the last few years which affects both job seekers and companies alike. Fraudulent companies and individuals lure job-seekers using multifarious methods on digital media platforms. Although several machine learning-based approaches exist for the automatic detection of fake job posts, they lack high accuracy and show skewed performance on imbalanced data. In addition, the influence of feature selection is not very well studied. This study overcomes these limitations using selective features through Chi-square and principal component analysis (PCA). The influence of dataset imbalance is also investigated through the synthetic minority oversampling technique (SMOTE). The performance of the proposed model is compared with individual machine learning models, as well as, existing state-of-the-art models. Results indicate that using SMOTE with Chi-square-based selective features yields the best results with a 0.99 accuracy using the proposed model. K-fold cross-validation further corroborates these results. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
13807501
Volume :
83
Issue :
6
Database :
Complementary Index
Journal :
Multimedia Tools & Applications
Publication Type :
Academic Journal
Accession number :
175166573
Full Text :
https://doi.org/10.1007/s11042-023-15173-8