MUMBAI, India, Jan. 7 -- Intellectual Property India has published a patent application (202511104242 A) filed by Lovely Professional University, Phagwara, Punjab, on Oct. 29, 2025, for 'machine learning based e-commerce fraud detection system.'
Inventor(s) include Preeti Khurana; Dr. Balraj Kumar; Dr. Sorabh Lakhanpal; and Navdeep Singh.
The application for the patent was published on Dec. 12, under issue no. 50/2025.
According to the abstract released by the Intellectual Property India: "A machine learning based e-commerce fraud detection system comprising a dataset generation module 101 configured to collect and categorize legitimate and fraudulent e-commerce websites into a dataset named ELFW-2031, a feature engineering module 102 extracts fifty features from six heterogeneous sources, including website content, social media, and technology analysis resources, a data preprocessing module 103 cleans, normalizes, and structures the data for model training, a model training module 104employs XGBoost gradient boosting protocols to train multiple classification models, a model evaluation module computes performance metrics, achieving F1 scores of 96.88% with all features and 96.53% using features independent of external services, a scalability optimization module ensures detection operations remain efficient without relying on external services."
Disclaimer: Curated by HT Syndication.
