Cross-lingual Text Selection and Simplification for Bilingual Education

Reading proficiency is a fundamental component of language competency. However, finding topical texts at an appropriate reading level for foreign and second language learners is a challenge for teachers. This task can be addressed with natural language processing technology to assess reading level and simplify text. This proposal presents a research plan for the development of tools for reading level assessment based on statistical language models. In addition, we propose adapting paraphrasing and summarization techniques for the task of text simplification. Coupled with an information retrieval system, these tools will be used to select and simplify reading material in multiple languages for use by language learners. A pilot study of the use of statistical language models for reading level assessment confirms that this approach is promising. Based on the large number of Spanish-speaking students learning English in the U.S., we selected English and Spanish as the specific languages for this work, but we expect the techniques developed will generalize to other languages as well.