MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641018874 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Feb. 19, for 'machine learning-driven energy complexity framework for predicting optimal data structures in workload-specific execution.'
Inventor(s) include Dr. R. Manjula; Shann Antony Suresh; and John Poly.
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: "Machine Learning-Driven Energy Complexity Framework for Predicting Optimal Data Structures in Workload-Specific Execution. The invention relates to a machine-learning-driven analytical framework for evaluating and predicting the energy complexity of data structure operations under workload-specific execution conditions. The framework integrates an analytical module that models operation-level characteristics with a machine-learning engine trained on labelled energy data to generate accurate energy predictions. By analysing workload parameters such as operation frequencies, access patterns, and dataset properties, the system predicts energy consumption using classification and regression models. Feature-importance analysis identifies the most influential workload attributes, while energy-per-operation scaling curves support prediction stability across varying dataset sizes. The decision component combines analytical descriptors and machine-learning outputs to recommend the data structure expected to yield the lowest energy consumption for the given workload. The framework continuously improves predictive accuracy by incorporating historical execution behaviour, enabling adaptive and energy-efficient data structure selection."
Disclaimer: Curated by HT Syndication.
