The Automated Text Adaptation Tool

Text adaptation is a teacher practice used to help with reading comprehension and English language skills development for English language learners (ELLs) (Carlo, August, McLaughlin, Snow, Dressler, Lippman, Lively, & White, 2004; Echevarria, Vogt and Short, 2004; Yano, Long and Ross, 1994). The practice of text adaptation involves a teacher’s modification of texts to make them more understandable, given a student’s reading level. Teacher adaptations include text summaries, vocabulary support (e.g., providing synonyms), and translation. It is a timeconsuming, but critical practice for K-12 teachers who teach ELLs, since reading-level appropriate texts are often hard to find. To this end, we have implemented the Automated Text Adaptation Tool v.1.0 (ATA v.1.0): an innovative, educational tool that automatically generates text adaptations similar to those teachers might create. We have also completed a teacher pilot study. Schwarm and Ostendorf (2005), and Heilman, Collins-Thompson, Callan, and Eskenazi (2006) describe related research addressing the development of NLP-based reading support tools. During our interactive demonstration, conference participants can (a) login to the Internet-accessible tool, (b) import text files, and (c) experiment with adaptation features. We are currently interested in feedback from the computational linguistics community to inform tool development related to (a) feature enhancement, and (b) ideas for new NLP-based features. Until now, our primary source of feedback has been from teachers toward tool development from an educational perspective.