Describes a method for automatically selecting key frames that can be used to represent the total image sequence of visual documents. An abstracting algorithm developed at the National Aeronautics and Space Administration is explained, and extensive examples of abstracted motion sequences are presented. (eight references) (LRW)

Proposes and explains a new mechanism for controlling the access requests in information protection systems that is based on Newton's interpolating polynomial. Access control matrices are discussed; single-key-lock systems are described; relevant algorithms are explained; and an example of the new mechanism is included. (10 references) (LRW)

Analyzes the amount of compression possible when arithmetic coding is used for text compression in conjunction with various input models. Algorithms are analyzed; modeling effects are considered; scaling is discussed; higher order models are examined, including prediction by partial matching; and coding effects are described. (34 references) (LRW)

Three experienced and eight novice physicians were asked to solve three problems. Results indicated that the ability to incorporate contextual information contributed to effective solutions. Experienced physicians tended to use more inclusive approaches, although both groups demonstrated premature diagnostic bias. (SK)

Two Statistical Analysis System (SAS) macros are presented that perform the modified principal components approach of L. R. Tucker (1966) to modeling generalized learning curves analysis up to a rotation of the components. Three SAS macros are described that rotate the factor patterns to have characteristics Tucker considered desirable. (SLD)

A simulation study was conducted to determine whether latent class model parameters are recovered adequately by Latent Transition Analysis (LTA). Results indicate that parameter recovery is satisfactory overall and that the benefits of adding indicators outweigh the costs. Additional indicators also improve standard errors. An example of LTA is…

Discusses problems of matching personal names in information systems and factors influencing design of personal name matching algorithms. Several examples of name matching systems in the fields of art history, bibliography, commerce, genealogy, and law enforcement used for the purposes of authority control, information retrieval, and duplicate…

A general approach is described for the analysis of categorical data when there are missing values on one or more observed variables. The method is based on generalized linear models with composite links. Situations in which the model can be used are described. (SLD)

Enriching factor rotation algorithms with the capacity to conduct repeated searches from random starting points can make the tendency to converge to optima that are merely local a way to catch rotations of the input factors that might otherwise elude discovery. Use of the HYBALL computer program is discussed. (SLD)

A method for hierarchical agglomerative polythetic (multivariate) clustering, based on unweighted pair group using arithmetic averages (UPGMA) is compared with the original beta-flexible technique, a weighted average method. Reasons the flexible UPGMA strategy is recommended are discussed, focusing on the ability to recover cluster structure over…

Presents two methods for including weights in distance-based nonlinear multivariate data analysis. One method assigns weights to the objects, while the other is concerned with differential weighing of groups of variables. Discusses applications of these weighting schemes and proposed an algorithm to minimize the corresponding loss function. (SLD)

Presents minimization and maximization models for parallel test construction under constraints. The minimization model constructs weakly and strongly parallel tests of minimum length, while the maximization model constructs weakly and strongly parallel tests with maximum test reliability. (Author/SLD)

Proposes an algorithm for the assembly of multiple test forms in which the multiple-form problem is reduced to a series of computationally less intensive two-form problems. Illustrates how the method can be implemented using 0-1 linear programming and gives two examples. (SLD)

Discussion of models for information retrieval focuses on an application of linear algebra to text clustering, namely, a metric for measuring cluster quality based on the theory that cluster quality is proportional to the number of terms that are disjoint across the clusters. Explains term-document matrices and clustering algorithms. (Author/LRW)

Presents a multilevel model of the information filtering process that permits document classification. Evaluates a document classification approach based on a supervised learning algorithm, measures the accuracy of the algorithm in a neural network that was trained to classify medical documents on cell biology, and discusses filtering…