AI-powered Data Classification
A public contracting authority faced the challenge of accurately classifying large volumes of structured data while ensuring that all information used in the data classification process was treated confidentially.
To this end, IT Sonix trained a machine-learning model that significantly reduces manual work, ensures data quality, and reliably supports legal requirements. The AI-based data processing rapidly accelerated processes at the client’s organization that had been in place for years.
Key Areas of Cooperation
Until now, classification had been performed entirely manually using the dual-control principle. This approach was time-consuming, costly, and difficult to scale. The client therefore wanted to integrate a cost-effective solution for AI-based data processing into its workflows that could be implemented as quickly as possible. The biggest challenge for the client was precisely distinguishing between corporate data and personal data.
This classification is essential for downstream processes, particularly with regard to data protection and anonymization requirements. The existing data came from various sources, was sometimes inconsistent, and was of only moderate quality. The goal of the collaboration with IT Sonix was therefore to automate the initial data classification without compromising the high level of technical and legal certainty. At the same time, the manual effort was to be reduced by at least half.
Key factors:
- High classification accuracy despite inconsistent data
- Integration into existing processes and infrastructure
- Machine Learning
- MLOps
- Private Cloud
Our Project KPIs
Results and Outlook
Initial data classification is now fully automated; human intervention is required only during the review phase. This significantly speeds up the overall process, reduces costs, and simultaneously improves the quality of the results. In terms of numbers, the AI achieves an accuracy of over 99 %, with an F1 score and recall of over 98 % each. In production, the quality is confirmed by a correction rate of less than 1 % during manual review.
The solution lays the groundwork for further automation steps. In the future, the model can be expanded to include additional data classes and used for other business processes in the public sector.
Our Services and Case Studies in Artificial Intelligence
- AI Consulting and Workshops,
- Implementations,
- AI-powered data classification,
- AI Document Management,
- AI-based process automation,
- AI agents,
- Machine Learning,
- LLMs
Artificial Intelligence Consulting
Sustainable know-how transfer for your company thanks to our AI consulting: workshops, use case evaluation, implementation consulting …
AI Document Management
LLM-Based Document Analysis and Processing in Healthcare …
AI-Based Process Automation
AI-based process automation for corporate data sets and data mapping in the public sector …
We'd be happy to advise you
Dr. Danny Hucke
We understand that every project is unique. Dr. Danny Hucke looks forward to speaking with you to find customized solutions.
