learningandteaching.PDF

An Overview of LSA-Based Systems for Supporting Learning and Teaching 1 Philippe DESSUS Laboratoire des Sciences de l’Education, University of Grenoble, France Abstract. We present an overview of LSA-based systems that have been used in instructional settings. Current research on this subject does not take into account the cognitive aspects of learning and teaching, and describes the systems at a technical level. We propose a cognitive-based classification of these systems that can lead to the design of novel LSA-based applications. Keywords. Latent Semantic Analysis, Tutoring Systems, Interactive Learning Environments, Natural Language Processing, Feedback, Learning, Teaching. Introduction Latent Semantic Analysis (LSA, [1]) is a well-known technique that captures semantic information in texts by uncovering word-usage regularities. Extensive research on LSA has proven its efficiency in the domain of natural language processing, and more specifically for computer-based instruction—tutoring systems, interactive learning environments [2]. The power of LSA lies in its versatility, resulting from its simple procedure: raw texts (representing different kinds of discourse, like teacher-student interactions, course texts, or student productions) are subject to a fast processing, as follows. Words that are shared by similar paragraphs, as well as paragraphs that contain similar words are represented similarly [3]. Without any complex human pre-processing, this mechanism allows semantic-related comparisons of words and texts (e.g., cohesion or meaning); and can lead to the simulation of the high levels of human cognition seen in instruc-