UnitedHealth Group has been granted a patent for a hybrid question-answering (QA) application that combines retrieval QA and deep QA. The system uses feedback from retrieval QA to enhance the deep QA application, improving accuracy in answering user queries. GlobalData’s report on UnitedHealth Group gives a 360-degree view of the company including its patenting strategy. Buy the report here.

According to GlobalData’s company profile on UnitedHealth Group, Content recommendation models was a key innovation area identified from patents. UnitedHealth Group's grant share as of February 2024 was 42%. Grant share is based on the ratio of number of grants to total number of patents.

Hybrid question-answering application combining retrieval and deep learning

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

A recently granted patent (Publication Number: US11921761B2) discloses a computer-implemented method that involves receiving an input question from a user computing entity, generating a deep learning answer using a deep question answering (QA) application, and determining the confidence score for the answer. If the confidence score does not meet a certain threshold, a retrieval QA application is queried for a retrieved answer with a high confidence score. The method also involves training an artificial neural network using training data generated from retrieved answers and expert user answers. The system aims to provide accurate answers by combining deep learning and retrieval QA techniques, ensuring high-quality training data for the neural network.

Furthermore, the patent describes a computing system that implements the method, utilizing a deep QA application with an artificial neural network to generate answers and a retrieval QA application to provide answers with high confidence scores. The system identifies relevant documents, extracts candidate answers, and compares them to the input question to determine the best answer. In cases where the confidence score threshold is not met, the system requests an expert user answer. The system also generates training data for the neural network based on retrieved answers, aiming to improve the accuracy of the deep learning answers. By combining deep learning and retrieval QA approaches, the system enhances the overall performance of the QA platform, ensuring accurate and reliable answers for user queries.

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

Premium Insights


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, 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.