Detection of OCR Quality on Patents
Project
The goal of this project is to linguistically enrich and semantically annotate patent documents by analyzing the patent document itself. This will assist in discovering hidden or embedded
information, making it explicit and associate it with the patent document in a (semi)-automatic
way. Additionally, the resulting system will allow for concept based searching over a wide range of patent annotations.
Project Partners
- University of Massachusetts Amherst
- Matrixware Information Services GmbH
