tailieunhanh - Báo cáo khoa học: "Simplifying Text for Language-Impaired Readers"

We download the original newspaper articles automatically from the WWW2, and apply a number of processing stages sequentially. Lexical Tagger The tagger (Elworthy, 1994) assigns and ranks part-of-speech (PoS) tags for each word in a sentence using a rst-order HMM. The tagger includes an unknown word guesser with an accuracy of around 85%, and a large diskresident lexicon specialised to newspaper text. Morphological Analyser The morphological analyser (an enhanced version of the GATE project lemmatiser (Cunningham et al., 1996)) is based on nite state techniques, and performs an accurate and e cient in ectional analysis of the words in. | In Proceedings of the 9th Conference of the European Chapter of the ACL EACL 99 Bergen Norway Simplifying Text for Language-Impaired Readers John Carroll Yvonne Canning Guido Minnen Siobhan Devlin Darren Pearce Cogniti e and Computing Sciences University of Sussex . Brighton BN1 9QH UK . Jo gidomi Automatic text . to language-impa d cadets . a unexplored area in natural leagues processing Wo describe a generic . m tor tex srmplilioatiou curmutly at the prototype stage incorporating ạ range ot state-of-the-art language processing tools. We are apply ng the system to help people with aphasia various languag impairment. typ rally e -curring as a result ot a stroke or head injury to understand English newspaper articles Aphasic people may encounter ma V problems 11 h d atel DTli 1999 Ura these problem an be ot a lea. nature .mce less frequent words ate often not readily . lab e. and . of . syntnehe nature that particular construct may pose sc ion to under standing In add ion to those general aspects of text them am .1 o 1 obi ms .pe-ciho to newspaper t xt for xample. the often eery compact summary-like 11 st paragraph in an art els long sentences ho use. of noun compounds and long soquenc of adje frees and frequent use of he pcsiee Although there is wide variation the language problems associated with aphasia depend ng on such factor as locus of b arn mjury. aphasia type and pr -aphasic literacy level many aphasic poopl would benoht from a . of the sort we describe . . Wo outline below the pro strategy of th system and th use -centered evaluation we mtend to carry out Wo envisage that the rnsult. of tin. project wall be of use not only to aphasic Individuals but . to other groups .uch a. m speakers preho .on of written Engl h fr edybrnftedfomlgn .ang ago .. This work is being carried out on the project PSET Practical Simplification of English Text fun del woth y UK - s refs. R .