Hybrid word/Part-of-Arabic-Word Language Models for arabic text document recognition.
Mohamed Faouzi BenZeghibaJérôme LouradourChristopher KermorvantPublished in: ICDAR (2015)
Keyphrases
- language model
- n gram
- handwritten words
- translation model
- handwritten documents
- word recognition
- text documents
- language modeling
- handwriting recognition
- text classification
- probabilistic model
- document representation
- statistical language modeling
- bag of words
- document retrieval
- multiword
- word segmentation
- character recognition
- text mining
- information retrieval
- keywords
- speech recognition
- co occurrence
- retrieval model
- object recognition
- topic models
- spoken term detection
- term frequency
- query terms
- test collection
- part of speech
- vector space model
- word level
- query expansion
- web pages