Challenge
When collaborating with partner companies, several thousand document must be shared every year. Besides information that is important for collaboration, these documents also include critical sections. These sections must not be shared for reasons of data protection or competition law. These sections must be identified and anonymized. This work is time-consuming and error-prone.
Solution
Using machine learning, critical sections can be quickly identified and automatically labeled. An additional manual check of the pre-labeled sections is possible. The AI used continuously learns from manual improvements.
Benefit
- Documents no longer have to be searched for critical text sections and "anonymized" by several clerks.
- Optimal support of clerks by recommended sections
- Continuous improvement of results by a learning AI
