Enhancing Authentic Web Pages for Language Learners

Abstract

Second language acquisition research since the 90s has emphasized the importance of supporting awareness of language categories and forms, and input enhancement techniques have been proposed to make target language features more salient for the learner. We present an NLP architecture and webbased implementation providing automatic visual input enhancement for web pages. Learners freely choose the web pages they want to read and the system displays an enhanced version of the pages. The current system supports visual input enhancement for several language patterns known to be problematic for English language learners, as well as fill-in-the-blank and clickable versions of such pages supporting some learner interaction. 1