Language Muse: Automated Linguistic Activity Generation for English Language Learners

Nitin Madnani, Jill Burstein, John Sabatini, Kietha Biggers, Slava Andreyev

Abstract: Current education standards in the U.S. require school students to read and understand complex texts from different subject areas (e.g., social studies). However, such texts usually contain figurative language, complex phrases and sentences, as well as unfamiliar discourse relations. This may present an obstacle to students whose native language is not English — a growing sub-population in the US. 1 One way to help such students is to create classroom activities centered around linguistic elements found in subject area texts (DelliCarpini, 2008). We present a web-based tool that uses NLP algorithms to automatically generate customizable linguistic activities that are grounded in language learning research.