MUMBAI, India, March 13 -- Intellectual Property India has published a patent application (202411063200 A) filed by Indian Institute Of Technology, Jodhpur, Rajasthan, on Aug. 21, 2024, for 'machine learning based prediction of value of assets.'
Inventor(s) include Upadhyay, Vishvajeet; and Paul, Angshuman.
The application for the patent was published on March 13, under issue no. 11/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to methods and systems (102) for predicting value of assets in a market. The system (102) comprises a training data preprocessing module (210) to segregate historical market data (222) into first, second, and third phase 5 datasets. A machine-learning model ensemble (MLME) module (212) is trained to generate a predicted value of the asset using three predictive models trained on the respective phase datasets. A market phase determination module (214) is configured to determine the current market phase based on recent market data (226) and an adaptability module (216) assigns weightages based on determined current market 10 phase to each predictive model which are fine-tuned using synthetic data (228). The predicted value is based on these assigned weightages, allowing the system to adapt to changing market conditions and improving the prediction accuracy of the asset value."
Disclaimer: Curated by HT Syndication.
