Please look at work being done at the University of Manchester. GATE is a NLP toolkit that contains a user interface that allows users to find named entities in unstructured text documents. These entities include names of person, organizations, and geographical locations.
The first release of GATE was over 15 years ago and has been open source for the entire time it was released. People working in the NLP industry for some time should be very familiar with GATE.
Here is the overview page on their site. http://gate.ac.uk/overview.html
Furthermore, a Google search on "named entity extraction" returns many links.
The first link is a link to Wikipedia. On this link they mention GATE as an example of software used to perform named entity extraction. Here is the link to that Wikipedia page:
This capability is the hardest part of this patent. The subsequent claims on relating like entities across several documents is trivial once named entity detection is available.