Modernizing historical Slovene words with character-based SMT.
Yves ScherrerTomaz ErjavecPublished in: BSNLP@ACL (2013)
Keyphrases
- historical manuscripts
- text recognition
- chinese characters
- keywords
- text documents
- historical documents
- word alignment
- historical data
- english words
- optical character recognition
- related words
- word recognition
- n gram
- statistical machine translation
- word spotting
- writing style
- writing styles
- handwritten words
- text classification
- unknown words
- neural network
- word segmentation
- information retrieval