Text Understanding Facilitator Framework

A Word document having 20 pages, for example an insurance contract, is neither easy to use for a human nor for a machine. They both profit from enriching such documents with additional meta data, either derived externally (e.g., classification) or obtained from the unstructured document. Documents thus become more discoverable, comparable and machine-readable.

TUFF is our comprehensive framework for enriching arbitrary text documents. In the demo, you can test the following features:

• Automatic extraction of core concepts
• Summarization of text passages
• Extraction and highlighting of essential terms and entities

Turbocharge your text documents

The use cases for this framework are innumerable. Even private users can benefit, for example, from the ability to see key terms in industry news or other publications at a glance and quickly jump to the most relevant passages.

In addition, this technology enables, among many other things:

• Digital forensics, for instance compilation of case-related communication threads
• Real-time monitoring of communication and data, for instance for compliance or data protection purposes (GDPR)
• Automatic landing pages: re-composition of existing website content based on the extracted concepts