Automatic accurate broken character restoration for patrimonial documents.
Bénédicte AllierNadia BaliHubert EmptozPublished in: Int. J. Document Anal. Recognit. (2006)
Keyphrases
- document collections
- document retrieval
- information retrieval
- metadata
- web documents
- legal documents
- database
- relevant documents
- xml documents
- image restoration
- document clustering
- digital libraries
- semantic information
- semi automatic
- keywords
- text documents
- high quality
- highly accurate
- document classification
- vector space model
- printed documents
- multimedia
- relevance feedback
- document content
- document representation
- ranked list
- computationally efficient