Information Retrieval List Digest 046 (February 4, 1991) URL = http://hegel.lib.ncsu.edu/stacks/serials/irld/irld-046 IRLIST Digest February 4, 1991 Volume VIII, Number 3 Issue 46 ********************************************************** III. JOB ANNOUNCEMENTS A. Research Literature Analyst University of Washington IV. PROJECT WORK B. Bibliographies 1. Selected IR-related dissertation abstracts ********************************************************** III. JOB ANNOUNCEMENTS III.A. Fr: JudithSchrier Re: Position Available: Research Literature Analyst Index scientific literature on nonhuman primates; produce topical bibliographies; perform custom subject searches. Expertise in neurobiology/neuroanatomy/neurophysiology is especially relevant. This position requires a Bachelor's degree in a biomedical or zoo- logical field, experience using computers, and 2 years professional experience in literature analysis. A graduate degree and experience reading scientific literature and doing subject indexing are pre- ferred. Official University of Washington applications (resumes may not be substituted) must be received by office closing or postmarked by mid- night, Thursday, February 28, 1991. To request or submit applications, contact the UW Staff Employment Office, 1320 N.E. Campus Parkway, Seattle, WA 98195 [206-685-2728], an equal employment, affirmative action employer. ********************************************************** IV. PROJECT WORK IV.B.1. Fr: Susanne Humphrey Re: Selected IR-related dissertation abstracts The following are citations selected by title and abstract as being related to Information Retrieval (IR), resulting from a computer search, using BRS Information Technologies, of the Dissertation Abstracts Online database produced by University Microfilms International (UMI). Included are UMI order number, title, author, degree, year, institution; number of pages, one or more Dissertation Abstracts International (DAI) subject descriptors chosen by the author, and abstract. Unless otherwise specified, paper or microform copies of dissertations may be ordered from University Microfilms International, Dissertation Copies, Post Office Box 1764, Ann Arbor, MI 48106; telephone for U.S. (except Michigan, Hawaii, Alaska): 1-800-521-3042, for Canada: 1-800-268-6090. Price lists and other ordering and shipping information are in the introduction to the published DAI. An alternate source for copies is sometimes provided. Dissertation titles and abstracts contained here are published with permission of University Microfilms International, publishers of Dissertation Abstracts International (copyright by University Microfilms International), and may not be reproduced without their prior permission. AN University Microfilms Order Number ADG90-28505. AU CORNELIO, ALOYSIUS. TI A STRUCTURE-FUNCTION MODELING PARADIGM FOR ENGINEERING DATABASES. IN The University of Florida Ph.D. 1989, 208 pages. DE Engineering, General. Computer Science. AB A unifying data representation paradigm is proposed for modeling and integrating engineering systems. This paradigm, called the structure-function paradigm, consists of structural information, functional information, and interactions between structures and functions. Structural information models the physical system, physical constraints, and the system state, whereas functional information models the overall behavior of the system. Well-defined interactions model the many-to-many semantic associations between the structural and the functional information. The structure-function paradigm uses object-oriented concepts where the structures are represented by structural objects, the functions are represented by functional objects, and the interactions between structures and functions are represented by interaction objects. This paradigm is based on the premise that structural and functional models of inherently complex natural and man-made systems are not isomorphic, i.e., the abstractions in the structural domain do not correspond to abstractions in the functional domain, and vice versa. Moreover, a function is carried out by many structures with possible alternate options; in addition, a structure will support many functions, also with alternatives. Therefore, structures and functions are represented and abstracted independently by structural object hierarchies and by functional object hierarchies, respectively, and the relationship between these hierarchies are modeled by interaction objects. These hierarchies are further extended to directed acyclic graphs. Invariants ensure that the structural objects and the functional objects are abstracted correctly. A set of rules for preserving the integrity and correctness of composite structural objects and composite functional objects are presented. The proposed representation also supports an active database environment by modeling events which monitor input data and system state, and initiate and terminate actions. These events trigger a set of constraints to ensure the consistency of the system representation, thus making this paradigm suitable for monitoring and simulating applications. The utility of this paradigm is demonstrated in two areas--discrete manufacturing and computer integrated manufacturing. In discrete manufacturing, the structure-function paradigm helps the designer in making design choices for building robotic workcells. In computer integrated manufacturing, the structure-function paradigm is used to integrate engineering databases with operational information such as application packages into one integrated environment. AN University Microfilms Order Number ADG90-18955. AU ROY, UTPAL. TI COMPUTER-AIDED REPRESENTATION AND ANALYSIS OF GEOMETRIC TOLERANCES. IN Purdue University Ph.D. 1989, 329 pages. DE Engineering, Industrial. AB Presently available CAD data models have no problem in handling the basic shape design of an object as required but they are not properly manufacturing-oriented. It is not possible to represent the dimensioning and tolerancing information or other technological information (such as material data, the surface roughness and accuracy data, feature information, etc.). In this thesis, we shall discuss the requirements of a new CAD data model and the suitability of adopting a feature-based representation scheme based on the hybrid CSG/B-Rep data structure. A front-end scheme necessary for interfacing the tolerance module with the solid model has been developed. This includes the development of (i) an efficient user interfacing system for inputing tolerance information, (ii) a suitable tolerance data module (to be attached to the solid model), and (iii) an information retrieval system from the stored database for tolerance analysis. Tolerance analysis has been carried out on two levels: individual part level and assembly level. Problems related to syntax and semantics of tolerance representation, consistency and validity of the specification (including the redundancy checking) are addressed in the part level analysis while the checking for the required mating conditions between different assembled parts is performed in the assembly level analysis. To facilitate the tolerance analysis in assembly, a generic assembly database is designed to encode explicit mating conditions between the parts for the establishment of functional relationships between the parts of the assembly. This database also helps in planning the assembly sequence. AN University Microfilms Order Number ADG90-22992. AU EL-HADDAD, FAHMI I. TI A REGIONAL DEVELOPMENT INFORMATION SYSTEM: A USER-ORIENTED DESIGN CONCEPT. IN The University of Nebraska - Lincoln Ph.D. 1989, 252 pages. DE Geography. Information Science. Urban and Regional Planning. AB This dissertation has explored, described and explained one means of linking the users and providers of geographic information applicable to the process of community economic development. This dissertation has produced a prototype design for a user-oriented information system for regional community economic development using personal computer technology. The research question for this dissertation asks: Is it possible to design a user-oriented community economic development information system which would effectively link data users with data sources on a regional basis. The basic primes is that a regional community economic development information system will enhance the ability of individuals, organizations, and governments to effectively utilize a process of identifying, acquiring, and employing geographic knowledge regarding the processes and phenomena which could effect the economic development of communities at the regional scale. The research design has produced an analysis of the characteristics of the information needs of those involved in regional community economic development, as well as the characteristics of the information available for such uses. The comparison between characteristics of users' needs and characteristics of information provided, has derived the basic structure and process of the suggested regional community economic development information system. This dissertation has shown that it is possible to create an opportunity to gain an improvement in knowledge, in the form of information, which should lead to an improvement in the ability of communities to understand their site and situation, to anticipate environmental change, and to achieve desired socio-economic conditions. Thus, this research suggests a means of enhancing a community's ability to persist and survive under changing economic conditions. This dissertation has also shown that information could be effectively utilized if such information is provided in a form that is appropriate to the needs and characteristics of the information users. AN University Microfilms Order Number ADG90-17133. AU LANTER, DAVID PHILLIP. TI TECHNIQUES AND METHOD OF SPATIAL DATABASE LINEAGE TRACING. IN University of South Carolina Ph.D. 1989, 219 pages. DE Geography. Computer Science. Library Science. AB A lineage-based graphical user interface enables a user to obtain information concerning the various thematic layers in a spatial database which contributed to any particular layer of interest. The component software parses input commands and determines if those commands to the spatial data processing and information systems are valid. The lineage information processor also creates a knowledge representation of the spatial database comprising a meta-database consisting of a semantic network that describes the various data layers in the spatial database and the relationship among these layers. The semantic network consists of parent and child links symbolizing the relationship among data layers, nodes describing the data layers in the spatial database, frames comprising attributes that describe the input data layers, the commands and command modifiers acting on those data layers, and characteristics of the final products. By means of rule-based processing, the lineage information program does not permit combinations of data layers that are incompatible or transformations that are non-valid. When such a problem is detected the lineage information program suggests commands to alter incompatible data layers so that the layers can be combined or transformed in the desired fashion. A graphical user interface enables a user to query the lineage information program concerning the lineage of data layers in the spatial database. AN University Microfilms Order Number ADG90-26315. AU SELF, PHYLLIS CHRISTINE KENNEDY. TI FORMAL AND INFORMAL COMMUNICATION PATTERNS AMONG AIDS RESEARCHERS: AN INVESTIGATION BASED ON COLLABORATION AND PRODUCTIVITY. IN University of Illinois at Urbana-Champaign Ph.D. 1990, 162 pages. DE History of Science. Information Science. AB The Acquired Immunodeficiency Syndrome (AIDS) has become a major health problem in the world and is discussed in the literature of many disciplines. Using bibliometric techniques and a mailed questionnaire, 87 international, AIDS researchers were studied in detail to determine whether or not patterns found in information communication resemble patterns found in formal communication. The database constructed for this study (AIDS Master File) was based on 8,926 AIDS-related articles downloaded from the National Library of Medicine's MEDLINE file for the years 1983 to 1987. Due to the size of the AIDS Master File a subset based on the work of authors who continued to publish AIDS-related articles for each of five years was created. Full and fractional productivity counts, collaboration, and author continuity in the publishing of AIDS-related articles were studied and contrasted with responses to the mailed questionnaire. The data reveal that the AIDS literature is written predominantly by the collaboration of 2 or more individuals. Only 3.5% were written anonymously or by a single individual. At the other extreme there was one article written by as many as 30 authors. Only 87 authors continued to publish for each of the 5 years. The authors under investigation collaborated in the formal literature with more unique individuals than expected. The range was from 12 to 507 collaborators over the 5 year period. The number of coauthors per paper ranged from 3.5 to 10.1 with a mean of 6.5. The range of full productivity was one individual contributing 177 articles to one individual contributing 5 articles with the mean at 24.1. A strong correlation was found to exist between fractional and full productivity rates for the 87 authors, r = 0.95, supporting Derek J. de Solla Price's prediction that the same researchers will be in much the same order regardless of the method use. There was also a strong correlation between productivity and collaboration for the 87 authors. The reluctance of AIDS researchers to respond to the questionnaire had a limiting effect on the study of informal communication. Findings from this portion of the study indicate that neither one's international collaboration activity nor length of time spent conducting AIDS research appears to relate to one's productivity. Most of the 87 researchers placed their work in the discipline of immunology followed by epidemiology and virology. The data suggest that a high degree of collaboration occurs in this research field through both formal and informal communication channels. However, data collected were not sufficiently complete to allow firm conclusions to be drawn on the relationships that might exist between the patterns of collaboration in authorship with the patterns of informal communication. AN University Microfilms Order Number ADGDX-88603. AU EGGHE, LEO. TI THE DUALITY OF INFORMETRIC SYSTEMS WITH APPLICATIONS TO THE EMPIRICAL LAWS. IN The City University (London) (United Kingdom) Ph.D. 1989, 202 pages. DE Information Science. AB Available from UMI in association with The British Library. This thesis is mainly concerned with the dual fundaments of informetrics. After an intuitive introductory chapter, we study, in a broad informetric context, general information production processes (IPP) (both discrete and continuous ones) and duality principles (between the sources and the items) in them in an exact and formalistic way. Classical informetrics evolves from this study as an example of a purely dual situation. The general duality technique is also able to recover new informetric laws (including a modelling of the Groos droop) that are easy to fit in practice. We present also parameter relations and classifications of some informetric laws, using only exact mathematical techniques. The most interesting features here are: the study of the group-free version of Bradford's law, the generalised Leimkuhler law and the place of Zipf's (or Pareto's) law in this context. Also the derivation of some formulae for some parameters, appearing in the group-dependent version of Bradford's law, is non-trivial and very useful in the sequel: they are e.g. basic tools in the fitting of the "nuclear" part of the Leimkuhler graph, even if a Groos droop is apparent, a result that has nice applications. Also the generalised Lotka and Leimkuhler functions are fitted. The thesis is rounded off by a summary of the results. AN University Microfilms Order Number ADGDX-89202. AU ELLIS, DAVID. TI THE DERIVATION OF A BEHAVIOURAL MODEL FOR INFORMATION RETRIEVAL SYSTEM DESIGN. IN University of Sheffield (United Kingdom) Ph.D. 1987, 340 pages. DE Information Science. AB Available from UMI in association with The British Library. This study outlines the derivation of a behavioural model intended to underpin thinking on questions concerning the design of information retrieval systems for academic social scientists. The historical background of information retrieval research is reviewed and the behavioural assumptions made in that research, characterised as the information retrieval model, are examined. Developments in the research tradition following from the early empirical tests on information retrieval systems are discussed, and problems experienced in the attempt to develop a coherent and practically useful research programme around the notion of relevance as a quantitative concept are analysed. A major alternative to work employing the information retrieval model has been the cognitive approach to information retrieval system design. In research so far undertaken in this approach the system has had to construct a cognitive model of the searcher's requirements in order for retrieval to take place. It is argued that the characteristics of social science information present particular difficulties for building such a model, and that the step of the system building a cognitive model of the searcher's requirements as a pre-requisite for retrieval is unnecessary. The features of an exploratory retrieval system for social scientists based on the characteristics of their information seeking patterns are outlined, the extent to which such features are available in existing systems, and the requirements for implementing the model on an operational system are considered. The argument of the study is that the behavioural approach, and the analysis of information seeking patterns into their characteristics, represents a coherent and practically useful basis for the design of exploratory information retrieval systems for academic social scientists. (Abstract shortened by UMI.). AN University Microfilms Order Number ADG90-20571. AU MCCUE, JANICE HELEN. TI ONLINE SEARCHING IN PUBLIC LIBRARIES: A COMPARATIVE STUDY OF PERFORMANCE. IN Columbia University D.L.S. 1987, 369 pages. DE Information Science. Library Science. AB The 1980s brought great emphasis on providing information resources to meet society's growing and changing needs. A major part of this movement was the growth of online database searching. The purpose of this exploratory study was to compare and analyze the quality of results of forty-two online database searches conducted in twenty-one public libraries located in sixteen states across the continental United States. Unobtrusive measurement was applied in the data collection process to obtain a true picture of the search service routinely offered at these twenty-one libraries from June 25, 1984, to October 24, 1984. An outside team of database search experts from leading United States financial institutions evaluated search results using a point system of scoring. The focus of the evaluation was the skill of the twenty-one online searchers in (1) retrieving citations that would provide information for a single multifaceted research question and (2) using two different databases. The experts affirmed that the best searches were attained by those searchers who used a broad search strategy thus retrieving a large number of citations. Regression analysis revealed that there is a strong relationship between the dependent variable quality and the independent variable quantity (the number of citations retrieved). Confirming the study's first hypothesis, the comparison of performance scores showed there was a wide difference in the quality of online searching from library to library. Addressing the study's second hypothesis that variability and quality of search performance was related to many factors, the only statistically significant variable that emerged from forty-two follow-up interviews was the monthly frequency with which the searchers perform searches. There are many issues within the sphere of quality online searching in public libraries that need answering. This study shows that the quality of the online search results in public libraries needs to be evaluated and measured with respect to the complete process such as the impact of (1) the nature of the topic, (2) the input with respect to the requestor, (3) the level of skill of the searcher, and (4) the input of the administrative practices. AN University Microfilms Order Number ADGDX-90011. AU WOOD-HARPER, A. T. C. TI COMPARISON OF INFORMATION SYSTEMS DEFINITION METHODOLOGIES: AN ACTION RESEARCH, MULTIVIEW PERSPECTIVE. IN University of East Anglia (United Kingdom) Ph.D. 1989, 246 pages. DE Information Science. AB Available from UMI in association with The British Library. This study identifies the confusion in the choice of methodologies which support the definition of an Information System prior to development and installation. In order to alleviate this confusion: (i) A comparison of these methodologies is made and concludes: that no single approach is best or sufficient; that any approach should support organisation activities and handle the data analysis perspective; and, finally that any method should make a contextual analysis of the environment. Consequently, this insight was used to identify 'generic' approaches and then they were classified. After this classification a synthesis is made and this framework is expanded to a methodology called Multiview. (ii) This methodology is then used for fieldwork consisting of six action research cases to accumulate lessons: the problem situation and the initial terms of reference; the problem solving team; action; methodology products and processes; and, the theory of using the Multiview methodology for defining an Information System. (iii) Finally, these lessons are used to make conclusions: that the methodology needed improvements; in defining an Information System there is a minimum set of contingencies consisting of 'problem solver', the situation and the methodology; and also that the definition process can be considered as a social process which can be explained by social theories. From the above conclusions recommendations are then made: to improve the Multiview methodology; to research the contingencies; and, finally to formulate a framework for the 'problem solver' to create a unique method relevant to the situation. (Abstract shortened by UMI.). AN University Microfilms Order Number ADG90-25727. AU KHAWAM, YVES JEAN. TI CITATION PATTERNS IN THE ARTIFICIAL INTELLIGENCE JOURNAL LITERATURE: A STUDY IN THE DETERMINATION OF AN INTERDISCIPLINARY RESEARCH BASE. IN The University of Wisconsin - Madison Ph.D. 1990, 153 pages. DE Library Science. AB This study proposed to determine whether the disciplines contributing to the journal literature of artificial intelligence (AI) share a common AI research base, or whether their AI research bases are discipline-specific. More precisely, the patterns in which three AI contributing fields (the sciences, social sciences, and humanities) cite themselves and others have been cross-analyzed so as to determine: (1) the differences among the three fields regarding the characteristics of each field's research base and focus within the AI journal literature, and (2) the extent to which that intersecting interdisciplinary area is employed by AI researchers. Hypotheses were analyzed across the three fields on the dependent variables of: field cross-citation rate, field cross-citation diversity, journal citation rate, proceeding citation rate, and age of cited literature, and the interdisciplinary AI research base was mapped out. Principal findings included: (1) the sciences tended to be the most parochial of the three fields in their use of AI literature, and (2) the idea of AI consisting of a completely interdisciplinary endeavor was refuted since the area candidate for an interdisciplinary research base resulted in a maximum extent of 0.1 percent of the research base proper to each field. The implication of such findings is that the overall mission of AI, the movement toward the creation of artificial intelligence, may be undermined by allowing disciplinary fragmentation to cloud the long-term conceptual goals. This study, however, makes no attempt to address this implication, but lays down the ground for it in terms of elucidating the various characteristics of the AI research bases. AN University Microfilms Order Number ADG90-20579. AU MURPHY, CATHERINE. TI MICROCOMPUTER STAND-ALONE ONLINE PUBLIC ACCESS CATALOGS: PRACTICES AND ATTITUDES OF SCHOOL LIBRARY MEDIA SPECIALISTS TOWARD STANDARDIZATION. IN Columbia University D.L.S. 1987, 207 pages. DE Library Science. Information Science. AB This study explored the state-of-the-art of microcomputer OPAC development in school library media centers in the spring of 1985. In particular, the study participants' bibligraphic practices and attitudes toward mainstream automation standards were investigated. The researcher was also interested in learning if the group endorsed special cataloging treatment for children's and curriculum materials. The researcher identified 161 OPAC systems in school library media centers in the United States and Canada. Ninety-six respondents using sixteen different programs, including three in-house systems, were included in the study. Most of the systems were determined by the researcher to be nonstandard in design. About ten percent of the OPACS included in the study did provide MARC formats, online authority control, or at least some of the recommended features. The study revealed that the majority of participants were not aware of mainstream library standards nor did they practice them in developing the OPACS in their school library media centers. The participants had a fairly agreeable attitude toward standards but were not as positive about traditional subject headings and children's cataloging. The strongest influence on the participants' adherence to standards as well as a positive attitude toward them, was the OPAC system used by the participant. Those who developed the more standard OPACS were more conforming. The professional activity or continuing education of the respondents, and most of the institutional variables, were not influential in a positive attitude. Experience and an instructional goal for the OPAC showed slight significance related to attitude and the larger, more urban school districts were more likely to purchase an OPAC that was standard in design. Most of the OPACS were developed at one site in a suburban school district. ********************************************************** IRLIST Digest is distributed from the University of California, Division of Library Automation, 300 Lakeside Drive, Oakland, CA. 94612-3550. Send subscription requests to: LISTSERV@UCCVMA.BITNET Send submissions to IRLIST to: IR-L@UCCVMA.BITNET Editorial Staff: Clifford Lynch lynch@postgres.berkeley.edu calur@uccmvsa.bitnet Mary Engle engle@cmsa.berkeley.edu meeur@uccmvsa.bitnet Nancy Gusack ncgur@uccmvsa.bitnet The IRLIST Archives will be set up for anonymous FTP, and the address will be announced in future issues. These files are not to be sold or used for commercial purposes. Contact Mary Engle or Nancy Gusack for more information on IRLIST. The opinions expressed in IRLIST do not represent those of the editors or the University of California. Authors assume full responsibility for the contents of their submissions to IRLIST.