The key concept behind Patent Reader is very simple. It reads the full content of the search document and using forensic linguistic techniques compares the terms it finds with the content of every patent or other document in the dataset. This a very different approach to traditional searching, which relies on keywords in combination or proximity, or semantic searching which expands the search to related concepts that occur in similar contexts.
With our system, a user does not need to perform multiple searches using a series of keywords; instead the program builds its own collection of terminology from the search document text. By searching on everything, the most relevant information is ranked highly, while distantly related material is automatically ranked lower, and unrelated material does not appear at all.