DiaLex: A Benchmark for Evaluating Multidialectal Arabic Word Embeddings.
Muhammad Abdul-MageedShady ElbassuoniJad DoughmanAbdelRahim A. ElmadanyEl Moatez Billah NagoudiYorgo ZoughbyAhmad ShaherIskander GabaAhmed HelalMohammed ElrazzazPublished in: WANLP (2021)
Keyphrases
- handwritten words
- unknown words
- co occurrence
- handwritten documents
- morphological analysis
- keywords
- printed documents
- euclidean space
- word recognition
- arabic documents
- printed text
- arabic language
- data sets
- word sense disambiguation
- vector space
- n gram
- word sense
- statistical machine translation
- noun phrases
- word level
- text classification
- natural language processing
- similarity measure
- writer independent
- compound words
- information retrieval