Learning to Summarize and Summarizing for Learning: Some Computer-Based Supports

Abstract

How computers can assess learning through course summaries written by students? Do they deliver reliable enough information for students to learn and sufficiently close to that of teachers? What possible research paths can be sketched from current research in this domain? I will try to give some answers to these questions in a fourfold talk. I will present firstly the links between learning and summarizing, and how LSA (Latent Semantic Analysis, Landauer et al., 2002) is a good model for accounting these both processes. Then I will show a comprehensive view of the research literature on LSA-based educational environments, detailing the different kinds of modules they embed (e.g., text selection/production, essay assessment, understanding/knowledge assessment, self-regulative learning/intentions assessment). Third, two interactive learning environments will be presented as examples: Apex 2 (Dessus & Lemaire, 2002), a system that provides course texts to be read and learned through summarization; and Resum’Web (Lemaire et al., 2005; Mandin et al., 2007), a system that detects the macro-rules used in summarizing texts and delivers feedback for revision purposes accordingly. Finally, likely research paths will be presented and discussed with the help of the attendants of this talk.