Artificial intelligence retrieves information from the deep web

High-quality information is often stored in databases that are only accessible via the portals of the respective providers. Such data is essentially invisible to conventional search engines. This part of the internet is known as the deep web, as the Chur University of Applied Sciences (HTW Chur) explained in a press release. Typical examples of such sources include electronic database, libraries and archives, online catalogues and a platform of the World Health Organization (WHO) that collects information on clinical studies. It often includes commercially relevant information.

Researchers at HTW Chur have now developed a solution for the automatic extraction of information from the deep web. This is able to access the WHO platform, for instance, which was especially relevant for researchers in the context of this project.

The new solution is based on artificial intelligence and can guesstimate which information might be most valuable for users. Having found the information, the system can then link it with the existing profiles of companies or active agents. In this way, it can also identify all current licensing agreements of biotech and pharma companies as well as potential changes in their key staff. It is hoped the system will enable decision makers to quickly gain a comprehensive and up-to-date picture of developments in the biotech and pharma sectors. They will furthermore be able to search for specific companies and active agents.

The researchers developed their solution during a two-year project, with funding from Innosuisse. Industry partner Venture Valuation is now offering the solution to its customers.