Nasdaq has been granted a patent for a computer system that selects feature sets based on metagradient information from machine learning processes. The system iterates through feature selection, training, metagradient calculation, and weight adjustment until convergence is achieved. GlobalData’s report on Nasdaq gives a 360-degree view of the company including its patenting strategy. Buy the report here.

According to GlobalData’s company profile on Nasdaq, AI for workflow management was a key innovation area identified from patents. Nasdaq's grant share as of January 2024 was 49%. Grant share is based on the ratio of number of grants to total number of patents.

Feature selection using machine learning metagradient information

Source: United States Patent and Trademark Office (USPTO). Credit: Nasdaq Inc

A computer system patent (Publication Number: US11861510B2) has been granted for a method that involves generating feature sets by selecting features from datasets based on assigned weights, performing a training process to create a model, calculating metagradient data to evaluate performance, adjusting weights based on the data, and repeating the process until convergence criteria are met. The system utilizes a hierarchical graph data structure to organize datasets and features, with operations including generating engineered datasets, applying penalty or constraint functions for feature selection, and calculating relevancy values for datasets to influence feature set generation.

The patent also covers a method performed on a computer system and a non-transitory computer-readable storage medium storing executable instructions for the same process. The method involves storing datasets and weights, generating feature sets based on weights, training models, calculating metagradient data, adjusting weights, and iterating the process until convergence. The computer-readable storage medium includes instructions for the same operations, emphasizing the importance of selecting features based on weights and utilizing a hierarchical graph data structure for dataset and feature organization. The weight assigned to a feature corresponds to its probability of selection within the dataset, and the feature set composition changes across iterations to optimize performance for the objective function used in each iteration.

To know more about GlobalData’s detailed insights on Nasdaq, buy the report here.

Data Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.