Post-instrument covariates are often included as controls in instrumental variable (IV) analyses to address a violation of the exclusion restriction. However, we show that such analyses are subject to biases unless strong assumptions hold. Using linear constant-effects models, we present asymptotic bias formulas for three estimators (with and…

Proposes that document space and query space have different structures in information retrieval and discusses similarity measures, term independence, and linear structure. Examples are given using the retrieval functions of dot-product, the cosine measure, the coefficient of Jaccard, and the overlap function. (Contains 28 references.) (LRW)

Describes an empirical Bayes method for predicting future use of library materials. Data from the University of Saskatchewan are used to illustrate practical applications of the model for simple prediction, prediction profiles, and prediction intervals. (21 references) (MES)

Reviews the literature on Bradford's law (i.e., an empirical relationship describing the distribution of scholarly articles in relevant journals). A model for the distribution of articles, based on random partitioning, is presented; and a mathematical formulation for a law of scattering is derived. The relationship of the model to Bradford's law…

Illustrates that the Poisson-lognormal model provides good fits to a diverse set of distributions commonly studied in bibliometrics and scientometrics. Topics discussed include applications to the empirical data sets related to the laws of Lotka, Bradford, and Zipf; causal processes that could generate lognormal distributions; and implications for…

Discusses distributed information retrieval systems that take into account the weights of descriptors from thesauri. Topics addressed include a mathematical model for information retrieval subsystems; organization of inverted files; models for the distributed homogeneous information systems; a distributed information retrieval system based on…

Discusses the generalized inverse Gaussian-Poisson (GIGP) process for informetric modeling. Negative binomial distribution is discussed, construction of the GIGP process is explained, zero-truncated GIGP is considered, and applications of the process with journals, library circulation statistics, and database index terms are described. (50…

Describes an evaluation model for expansion systems in information retrieval, i.e., systems expanding a user selection of documents to provide the user with a larger set of documents sharing the same or related characteristics. Highlights include a test protocol and evaluation measures for the efficiency of expansion systems. (Contains 29…

Discusses precision and recall in information science and proposes a new model based on fractal geometry for clusters of relevant documents. Search strategies for retrieving a group of relevant documents are reviewed; fractal sets and chaotic processes are described; and the new model is explained. (Contains 43 references.) (LRW)

Identifies a set of objects that is genuinely indigenous to management information systems and constructs a simple model based on them. The model's four basic entities, which define the scope of information theory, are data, information, knowledge, and wisdom. (Contains 46 references.) (KRN)

Proposes a new learning algorithm to improve the retrieval effectiveness of the search system used in the hypertext environment using an extended Boolean model with links to improve the ranking of retrieved items. Highlights include the basic retrieval scheme, the learning scheme, and a review of basic probabilistic retrieval models. (Contains 32…

Proposes a model for technical journal deselection based on the Bradford law of distribution. The model's operational prescriptives, characteristics of the Bradford distribution in the field of pediatrics, and relevance to collection management are discussed. (42 references) (KRN)

Examines the deductive approach to error diagnosis for intelligent tutoring systems. Topics covered include the principles of the deductive approach to diagnosis; domain-specific heuristics to solve the problem of generalizing error patterns; and deductive diagnosis and the hypertext-based learning environment. (Contains 26 references.) (JLB)

Presents a method of modeling the relevance relationship in information retrieval to answer the question of the theoretical limits of certain statistical methods. Hypergeometric probability distribution is used to construct an abstract model of a database of MEDLINE records, and results of tests of vector retrieval methods are reported. (28…

Describes research that was conducted to determine the search state patterns through which users retrieve information in hypertext systems. Use of the Markov model to describe users' search behavior is discussed, and search patterns of different user groups were studied by comparing transition probability matrices. (Contains 25 references.) (LRW)