This paper describes a knowledge-based strategy for recommending educational resources-worked problems, exercises, quiz questions, and lecture notes-to learners in the first two courses in the introductory sequence of a computer science major (CS1 and CS2). The goal of the recommendation strategy is to provide support for personalized access to…

Discusses latent semantic analysis (LSA), a statistical model for representing word usage in written language, and describes two experiments that were conducted with undergraduates to determine what aspect of knowledge, conceptual or factual, is being reflected in an LSA output from student essays. (Contains 21 references.)u (Author/LRW)

Discusses automated information retrieval, focusing on statistical patterns expected of a pair of terms that are semantically related to each other, guided by text generation conceptualization. Examines how the tendency of a content bearing term to clump, as quantified by previously developed measures of term clumping, is influenced by the…

Discusses the complicating effects of ambivalence on the validity of science attitude measurement scales. Cites studies indicating that the positive and negative effects about a particular psychological object are not necessarily bipolar. Suggests the use of factor analysis to check dimensionality in instruments. (TW)

Compares the success of lexical statistics, content analysis, and function words in determining the true author of "The Federalist." The function word approach proved most successful in attributing the papers to James Madison. Lexical statistics contributed nothing, while content analytic measures resulted in some success. (MJP)