Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{200724,
        author = {K Gopi chandu and G Prudhvinath and D Sharath kumar reddy and M Kanimozhi},
        title = {“Design and Implementation of an AI-Driven Fake Job Posting Detection System Using Naive Bayes Classification and TF-IDF–Based Natural Language Processing”},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {1463-1468},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=200724},
        abstract = {The rapid growth of online recruitment platforms has significantly increased employment opportunities for job seekers Online employment fraud has emerged as a significant societal and economic threat, with fraudulent job postings defrauding millions of job seekers annually. This paper presents Job Guard, an AI-driven fake job posting detection system that combines Multinomial Naïve Bayes (MNB) classification with Term Frequency–Inverse Document Frequency (TF-IDF) vectorization for accurate, real-time discrimination between genuine and fraudulent employment advertisements. The proposed system is trained and evaluated on the Employment Scam Awareness Corpus and Dataset (EMSCAD), comprising 17,880 job postings with a natural class imbalance of 88:12 (real: fake). The system achieves 96.1% accuracy, 95.4% precision, 94.9% recall, a 95.1% F1-score, and an AUC of 0.981, outperforming SVM, Random Forest, Logistic Regression, and LSTM baselines in both performance and computational efficiency. Notably, MNB + TF-IDF completes training in under 5 seconds, making it suitable for real-time deployment at scale. Ablation experiments confirm that bigram features and numeric attribute enrichment contribute meaningfully to detection performance. The paper also discusses SMOTE-based class imbalance handling, model interpretability via feature importance analysis, and deployment considerations for integration into online job platforms.},
        keywords = {Fake job detection, Naïve Bayes, TF-IDF, NLP, E employment fraud, Text classification, EMSCAD, Machine learning, Online fraud detection, Class imbalance.},
        month = {May},
        }