MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641108623 A) filed by Geethanjali College Of Engineering And Technology, Hyderabad on September 10, 2026, for Ai Powered Fake Job Posting Detection System.
Inventors include Mrs. T. Shailaja; Dr. K. Ambedkar; K. Manisha; V. Sanjana; K. Harshitha; and V. Sree Vaishnavi.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to an AI Powered Fake Job Posting Detection System designed to automatically identify potentially fraudulent, deceptive, or suspicious employment advertisements using Artificial Intelligence, Natural Language Processing, machine learning, anomaly detection, information verification, and explainable AI techniques. The proposed system receives job postings from authorized online recruitment platforms, company career websites, professional networking platforms, recruitment applications, social-media sources, uploaded documents, URLs, databases, and other permitted sources. The acquired job advertisements are processed to extract textual, structural, company, recruiter, contact, URL, salary, requirement, and metadata characteristics. The proposed system employs multiple complementary detection mechanisms to evaluate the authenticity and risk characteristics of job postings. Natural Language Processing analyzes semantic and linguistic patterns, while machine-learning models classify postings based on learned characteristics of legitimate and fraudulent advertisements. Anomaly detection identifies unusual posting characteristics, and dedicated analysis components evaluate company information, recruiter contacts, application URLs, duplicate listings, and source consistency. The outputs of these components are combined into a composite fraud-risk score and corresponding classification such as Legitimate, Suspicious, or Fake/High Risk. The system further provides explainable detection results by identifying the factors contributing to a classification, including suspicious language, unrealistic compensation, inconsistent company information, unusual recruiter details, suspicious application links, duplicate advertisements, or abnormal posting characteristics. A conversational Job Safety Assistant enables users to request explanations and safety guidance, while monitoring, alerting, detection history, and continuous-learning components support ongoing adaptation to emerging fraudulent patterns. The proposed invention therefore provides an intelligent, multi-factor, explainable, and continuously improving approach for detecting fake job postings and helping job seekers make safer and more informed employment decisions.
Disclaimer: Curated by HT Syndication.
