Simulating Text Understanding for Educational Applications with Latent Semantic Analysis: Introduction to LSA

Thomas K Landauer and Joseph Psotka

Abstract

The following seven articles describe new educational tools that rely on a unique capability of computer programs to abstract knowledge relationships from vast quantities of text, and using this, to determine the similarity of knowledge expressed in two or more texts. The tools can make comparisons among instructional sources and expository student writing, and use the results to assess the quality of the student compositions, and guide and even tutor students to revise and improve their compositions. The pedagogical motive is simple: students need to learn - and learn how to learn - in a manner that allows them to express their knowledge in discourse, and they need to learn the important skills of verbal knowledge expression. This is particularly important for ill-structured knowledge domains, such as history, the social sciences, and everyday practical knowledge, where answers are not clearly prescribed, but remain ambiguous and constantly changing.

Full Paper (PDF)