Phonetic based sentence level rewriting of questions typed by dyslexic spellers in an information retrieval context

摘要: This paper introduces a method combining spell checking and phonetic interpretation in order to automatically rewrite questions typed by dyslexic spellers. The uses finite state automata framework. Dysorthographics refers incorrect word segmentation which usually causes classical spelling correc-tors fail. specificities of the information retrieval context are that flexion errors have no impact since sentences lemmatised filtered several hypothesis can be processed for one query. Our system is evaluated on collected with help an orthophonist. error rate falls from 60% 22% (falls 0% 43% sentences).