Document Meaning Comparator

Ref: 10811
Software that can match multiple text based documents, based on comparing their meaning intentions rather than common words and/or the same strings of text.

Key advantages
  • Matches text based on their meaning
  • Usable with a large collection of documents
  • Matches for high similarity of runs of texts or concepts
  • Provides for user defined concepts of interest

Background

When dealing with a number of similar documents that need to be compared and/or merged, the task can be very resource intensive and take significant time to complete, particularly for long documents. This is mainly due to the almost infinite number of ways a particular idea can be expressed in written text. While programs such as Microsoft Word have a compare function, this is based solely on specific differences in words and text.

The invention

A software system that uses a number of language and comparison algorithms so that it can be used to determine the level of similarity between one or more documents and identifies the similar segments.

Applications

Any situation where the meaning of large documents needs to be compared. Examples include:

  1. Tender response documents that are written in response to the tender document. The response parts can be readily matched with the correct segments of the original document.
  2. When preparing a composite document from a set of overlapping documents areas of text of similar meaning can be readily identified and duplications eliminated.
  3. An audit function to search for content similar to known content which is not exactly the same which may be required for certain regulatory compliance.

Principal inventors