MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641017413 A) filed by Sudhakar G; and Sri Sai Ranganathan Engineering College, Coimbatore, Tamil Nadu, on Feb. 17, for 'adaptive bug triaging using transformer-based learning models.'
Inventor(s) include Bhuvaneshwari K; Nithish M; Madhan. J; Thanush P; Rajesh R; Pavithra B; and Maheshwari S.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "This project presents an advanced intelligent bug triaging system leveraging state-of-the-art machine learning and natural language processing techniques to automate issue labelling and developer assignment in large-scale open-source software projects. Building on prior deep learning models, including Bidirectional LSTM and CNN-LSTM architectures, the proposed system integrates transformer-based methods such as BERT and GPT for enhanced semantic understanding and classification performance. The solution supports multi-label classification on complex repositories, uses robust data augmentation, and incorporates project-specific metadata for context-aware recommendations. Extensive benchmarking across diverse open-source repositories demonstrates significant gains in precision, recall, and F1-score compared to previous approaches, even in datasets with class imbalance or sparse contributor data. The system's modular design promotes scalability, adaptability, and integration with development workflows, enabling faster issue resolution and improved resource allocation. This work contributes to the advancement of automated software maintenance by providing a scalable, generalizable, and efficient framework for intelligent bug triaging applicable to real-world software engineering environments."
Disclaimer: Curated by HT Syndication.
