Provides an overview of current image retrieval systems. Describes the architecture of the SOFM (Self Organizing Feature Maps) based image retrieval system, discussing the system architecture and features. Introduces the Kohonen model, and describes the implementation details of SOFM computation and its learning algorithm. Presents a test example…

Reviews the progress of parallel computing in information retrieval (IR) and stresses the importance of the motivation in using parallel computing for text retrieval. Analyzes parallel IR systems using a classification defined by Rasmussen; describes retrieval models used in parallel information processing; and suggests areas of needed research.…

Presents new methods for compressing bit matrices in large information retrieval systems which exploit possible correlations between rows of words and columns of documents. Three encoding methods are tested and compared--Shannon-Fano, arithmetic, and Huffman--and an appendix discusses binomial coefficients. (20 references) (LRW)

Uses Kintsch's proposition-based construction-integration theory of discourse comprehension to detail the user coding operations that occur in each of the three subsystems (Perception, Comprehension, Application) in which users process an information retrieval systems (IRS) message. Describes an IRS device made up of two separate parts that enable…

Presents an approach for dynamic database integration and query processing in the absence of information about attribute correspondences and global IDs. Defines different types of equivalence conditions for the construction of global IDs. Proposes a strategy based on ranked role-sets that makes use of an automated semantic integration procedure…

Designed to minimize information overload in telecommunications systems, the formal model developed in this paper predicts the usefulness of a message based on the available message features, and may be used to rank messages by expected importance or economic worth. The assumptions of binary and two Poisson independent probabilistic distributions…

Research supported in part by a Public Health Service Grant to the New School of Social Research and a National Science Foundation Grant to Stanford University. (VM)

Presents an overview of feedback within the cybernetics and social frameworks. These feedback concepts are then compared with the interactive feedback concept evolving within the framework of information seeking and retrieving, based on their conceptualization of the feedback loop and notion of information. (Author/AEF)

Presents a single framework for conducting topic analysis that performs both topic identification and text segmentation. Key characteristics of the framework are: representing a topic by means of a cluster of words closely related to the topic; and employing a stochastic model, called a finite mixture model, to represent a word distribution within…

Proposes an information retrieval system based on a hybrid model consisting of an inductive learning and neural network system. Evaluates the system's responses to incomplete queries and inconsistent indexing. Query terms, discriminant descriptors, and the American Documentation Institute's query and document titles used in the evaluation are…

Surveys theoretical and practical issues associated with pictorial information retrieval. Concentrating on still and moving pictorial forms of the visual image, this paper focuses on indexing pictorial material and discusses four models of pictorial information retrieval corresponding with permutations of the verbal and visual modes for the…

The goal of problem-solving education should be to develop tools for thinking that will constitute a shell or framework of action procedures that can be applied on an ever-changing database. These tools come in the form of heuristics that can be modified and adapted to new situations. (Author/MLW)

Investigates a cognitive model which proposes to account for performance in formation of a goal, selection of an information category, extraction of the information, integration of the information, and recycling until the goal of locating information in documents is met. Indicates that overall performance was facilitated by the model. (JK)

The paper develops a suggested model of process space for the Input, Process, Product, and Impact (IPPI) model of vocational education. Part 1 presents the need; Part 2 develops a schema for classifying process variables (human, physical, and organizational); their interactive nature is reflected in a dual classification system which enables users…

The Latent Semantic Indexing and MatchPlus systems examine similar contexts in which words appear and create representational models that capture the similarity of meaning of terms and then use the representation for retrieval. Text Retrieval Conference experiments using these systems demonstrate the computational feasibility of using…