MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202631124628 A) filed by C. V. Raman Global University on October 17, 2026, for An Intelligent Deep Learning–based System And Method For Software Reliability, Defect, And Fault Prediction.
Inventors include Suneel Kumar Rath; and Dr. Madhusmita Sahu.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: An intelligent deep learning based system and method for software reliability, defect, and fault prediction is disclosed. The system acquires software metrics such as size, complexity, coupling, cohesion, change history, defect logs, and test execution data for a plurality of software modules. A preprocessing module normalizes the metrics, handles missing values, removes noise, and corrects class imbalance. A feature learning module comprising a plurality of deep learning layers automatically learns a deep feature representation without manual feature engineering. A prediction engine uses the shared representation to estimate a defect probability and a fault severity for each module. A decision support module computes a software reliability index from these outputs, classifies modules into high, medium, or low risk categories using predefined thresholds, and generates recommendations for testing prioritization, maintenance scheduling, and release decisions. The invention enables early, scalable, and simultaneous prediction of defects, faults, and reliability. Figure 2 accompanies the abstract.
Disclaimer: Curated by HT Syndication.
