Memory-based Morphological Analysis and Part-of-speech Tagging of Arabic

Abstract

We explore the application of memory-based learning to morphological analysis and part-of-speech tagging of written Arabic, based on data from the Arabic Treebank. Morphological analysis is performed as a letter-by-letter classification task. Classification is performed by the k-nearest neighbor algorithm. Each classification produces a trigram of position-bound operations, each encoding segmentation, part-of-speech information, and letter transformations. The overlapping operation trigrams generated on the basis of an input word are converted into a lattice, from which all morphological analyses of the word are generated. Part-of-speech tagging is carried out separately from the morphological analyzer. A memory-based modular tagger is developed with a subtagger for known words and one for unknown words. On words not seen in training, the morphological analyzer attains a peak F-score of 0.47, while the tagger produces 66.4% correct tags. On all words, including words seen in training, the combination assigns a correct part-of-speech tag and generates all morphological analyses to about 91% of word tokens in running text [ABSTRACT FROM AUTHOR]; Copyright of Arabic Computational Morphology is the property of Springer eBooks and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)