US Patent 10067986 Discovering entity information

A technique for acquiring information about entities includes receiving starting data including an entity name and/or email address, generating a URL (Uniform Resource Locator) from the starting data, and downloading content from a website at the generated URL. Downloaded content from the website is analyzed to generate a set of entity-specific information and a confidence score. The confidence score specifies a likelihood that the entity-specific information pertains to the same entity that was described in the starting data. Using the improved technique, persons are able to obtain information about entities, even small, private entities about which information online is sparse, along with a measure of quality of the information obtained.