Documents and content units are rarely written in a standardized language that always uses the same concepts and sentence structures to describe the same content. Similar meaning can be expressed in multiple ways e. g.:

• “beautiful-looking, esthetically appealing painting”
  vs.
• “cool picture.”

On the other hand, same words can mean totally different things e.g.:

• “can”=“be able”
  or
• “can”=“container”.

While this rarely constitutes a problem for human text understanding, it usually means an insuperable obstacle for a keyword-based machine search and text mining.

ABBYY Compreno is the unique technology of natural language texts processing.

Due to over 15+ years of linguistic research, ABBYY has developed the innovative approach to text analysis. That allows to get to the meaning of the text and bring machine analysis a step closer to human text processing. Technologies based on statistic algorithms, do not actually “know” anything about the language. Therefore, they only learn from the concept frequencies and co-occurrences in the text. ABBYY Compreno Technology on the other side possesses knowledge about meanings and their relationships in the natural language.

• Learn more about ABBYY Compreno key components
  

• Further Details can be found in the ABBYY Compreno Technology White paper!

Back to:

• NLP/Semantic Technologies
• Compreno Technology Platform