MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202541134040 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; C Sasthi Kumar; and Manimaran B, Coimbatore, Tamil Nadu, on Dec. 31, 2025, for 'spam mail classification using ai.'
Inventor(s) include C Sasthi Kumar; and Manimaran B.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "Spam mail classification has become an essential component of secure and efficient email communication systems due to the rapid increase in unsolicited and malicious emails. These unwanted messages not only cause user inconvenience but also pose significant cybersecurity threats, including phishing attacks, malware distribution, identity theft, and financial fraud. To address these challenges, this project presents an intelligent, automated spam classification system that leverages advanced Artificial Intelligence (AI) and Machine Learning (ML) techniques to accurately distinguish between legitimate and spam emails. The proposed system incorporates a robust end-to-end pipeline that includes data collection, email parsing, preprocessing, feature extraction, and model training. In the preprocessing phase, emails undergo text cleansing, stopword removal, normalization, and HTML content filtering to convert raw inputs into machine-processable formats. Feature extraction techniques-such as Bag-of-Words, Term Frequency-Inverse Document Frequency (TF-IDF), and N-gram modeling-are used to generate high-quality feature vectors that capture both the semantic and structural characteristics of email content. These vectors are then used to train machine learning algorithms including Naive Bayes, Support Vector Machines (SVM), and deep learning-based models. The system employs a training-testing dataset split to evaluate model accuracy, precision, recall, and overall performance. The real-time classification module processes incoming emails through the trained AI model, delivering a binary output: spam or legitimate. Emails identified as spam are automatically redirected to a designated spam or quarantine folder, while legitimate messages are forwarded to the user inbox. A feedback loop enables continuous learning by incorporating user-reported false positives and false negatives into the training dataset, enhancing long-term system accuracy and adaptability. Overall, the proposed AI-driven spam mail classification system significantly improves email security, enhances user experience, and offers a scalable, efficient solution for modern communication platforms."
Disclaimer: Curated by HT Syndication.
