Student, Text and Curriculum Modeling for Reader-specific Document Retrieval

In today's language-learning classrooms, all of the students in a class almost always have the same text to read. Although students have different reading levels, it is impractical for a single teacher to find unique texts matched to each student's abilities. The REAP system was developed to make the process of providing students with individualized texts practical. The texts come in the form of authentic documents retrieved from the Web, and the system tracks and assesses students’ knowledge as they use the system. The system is able to find documents that meet various and individualized criteria. In this paper, we describe our work on modeling lexical familiarity. In particular, we detail the approaches taken for modeling the student's vocabulary knowledge, the contents of documents in the corpus, and the components of the curriculum. We also address related work and future plans.