Get started
Work

Human-Centric AI Framework for Fraudulent Job Advertisement Detection using NLP and Machine Learning

Year: 2026
Type: conference-paper
Abstract: The fast development of online recruitment tools has fundamentally changed the way the world of recruitment is conducted by enhancing the ease of communication between the employer and the potential employee. Nevertheless, digitization of the hiring atmosphere has also brought a tremendous surge in fraudulent job posts with the intention of taking advantage of the job seekers due to victims of misleading recruitment procedures. Such fraud posts may include inaccurate descriptions, farfetched salary scales and solicitation of sensitive information. The investment under consideration in this paper is the adoption of a human-oriented artificial intelligence system that is aimed at detecting fraudulent employment on the digital jobsposting platforms. The suggested system combines both natural language processing and machine learning methods in which textual patterns are analysed on an advertisement , suspicious aspects are detected and recruiter credibility indicators. The preprocessing of the text is done to tokenization and normalization and vectorization, which transform the unstructured job descriptions into structured feature representations. The machine learning classification models are then left to simulate the figure of the legitimate job posts and the fraudulent ones. Besides the detection accuracy, the proposed framework focuses on transparency and user protection through the presentation of interpretable outputs that are able to assist job seekers to be aware of the potential risks. It is anticipated that the suggested solution will make online recruitment systems more robust in terms of mitigating risk and focusing on trust, transparency and user safety
DOI: 10.1109/icssit69151.2026.11656513
Language: English
Open Access status: hybrid
OpenAlex ID: W7204092629
FWCI:
Cites: 8
Cited by:
Related to: 10