USPTO Patent for System Extracting and Categorizing Online Business Information

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted a patent (US12585714B1) to 6SENSE INSIGHTS, INC. for a system and method to extract and categorize online business information. The system uses a fine-tuned BERT model for URL classification and an LLM for content processing and tag generation.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has issued patent US12585714B1 to 6SENSE INSIGHTS, INC. The patent covers a system and method for extracting and categorizing online business information. Key components include a web crawler, a classification model based on a fine-tuned BERT architecture to identify relevant URLs, a content extractor, and a large language model (LLM) that processes extracted content to generate tags using custom prompts. This technology aims to improve the efficiency, scalability, and accuracy of automated business intelligence gathering from diverse online sources.

This patent grant is primarily an intellectual property matter and does not impose direct regulatory obligations on businesses. However, companies developing or utilizing similar AI-driven information extraction and categorization technologies should be aware of this patent. It may impact their freedom to operate or necessitate licensing agreements if their systems infringe upon the patented claims. The filing date was January 15, 2025, and the patent was granted on March 24, 2026.

Source document (simplified)

← USPTO Patent Grants

System and method for extracting and categorizing information from online sources

Grant US12585714B1 Kind: B1 Mar 24, 2026

Assignee

6SENSE INSIGHTS, INC.

Inventors

Ernest Kirubakaran Selvaraj, Samira Golsefid, Viral Tarun Bajaria, Satish Arjun Chilloji, Akshay Rajendra Shah, Amresh Sekar, Shubham Kumar Sunwalka

Abstract

A system and method for efficiently extracting and categorizing business information from online sources is disclosed. The system comprises a web crawler that obtains company domains from a database and collects depth-1 URLs from company homepages. A classification model, utilizing a fine-tuned BERT architecture, predicts which URLs contain relevant information for generating tags. A content extractor then extracts content from these predicted URLs using one or more modules. Finally, a large language model (LLM) processes the extracted content and generates tags using custom prompts designed for each tag category. These prompts are tailored to the nature of the extracted content, enhancing the context provided to the LLM. This multi-stage approach addresses challenges in processing large-scale, unstructured business data from diverse web sources, potentially offering improved efficiency, scalability, and accuracy in automated business intelligence gathering.