IRList Digest Sunday, 28 August 1988 Volume 4 : Issue 48 Today's Topics: Call for Papers - Panels for IJCAI '89 Abstracts - Dissertations selected by S. Humphrey [Part 5 of 5] News addresses are Internet: fox@fox.cs.vt.edu or fox%fox.cs.vt.edu@dcssvx.cc.vt.edu BITNET: foxea@vtcc1.bitnet (replaces foxea@vtvax3) ---------------------------------------------------------------------- Date: Thu, 18 Aug 88 09:15:45 EDT From: schmolze%cs.tufts.edu@RELAY.CS.NET Subject: Call for Panels for IJCAI-89 The IJCAI committee requests the submission of proposals for panel sessions to be presented at IJCAI-89. A panel session allows from three to five people to present their views and/or results on a common theme, issue or question. The panel topic must be both relevant and interesting to the AI community. The panel members must have substantive experience with the topic. However, the members need not be members of the AI community. Preference will be given to panels that demonstrate broad, preferably international, participation. A panel topic must be specified clearly and narrowly so it can be adequately addressed in a single session. Panel sessions run for 75 minutes. The format usually consists of an introduction by the chairperson with the purpose of providing the audience with a background for the ensuing discussion. The panel members, including possibly the chairperson, then present their views and/or results, followed by interchange between the participants and, finally, by interchange between the panelists and the audience. Preferably, the session ends with an overview by the chairperson. Panels may primarily serve to present information on a specific topic, such as recent important results or the status of important projects. Panels may focus on alternative approaches or views to a common question, where panelists present their approaches or views and the results they produced. Also, panels may be critical, where some members present an approach or view and other members criticize them, allowing time for rebuttals. REQUIREMENTS FOR SUBMISSION A proposal consists of a cover page, an overall summary and a summary of each member's presentation. The cover page should contain the following. o At the top of the first page, write "PANEL PROPOSAL". o Title of panel: The length should be similar to the lengths of titles of papers. o Chairperson: Name, affiliation, phone number, postal mailing address and electronic mailing address. Please give phone number and address for correspondence from the United States. o Members: Names, affiliations, phone numbers, postal mailing addresses and electronic mailing addresses. Please give phone numbers and addresses for correspondence from the United States. The overall summary should be brief, giving a clear description of the panel topic such that members of the general AI community can understand and appreciate it. It should explain how the member's presentations will be integrated. In addition, it should address the following questions. o What is the relevance and/or significance of the panel, including both the topic and the members? o What is the general AI interest in the topic? Please give evidence, such as recent important papers, workshops, etc. o How does the panel membership demonstrate broad, preferably international, participation? If it does not, why is narrow participation preferable? o If your topic has been discussed by another panel in a recent national or international AI conference, how will your panel differ from it? The overall summary should be from 500 to 1000 words in length. The final part of the proposal should be a brief summary of each member's presentation. This includes the chairperson if she or he will give a presentation. Each such summary should give a clear description of the member's view or approach, summarize results if appropriate, and demonstrate the connections to the panel topic. Where appropriate, each summary should support the arguments given in the overall summary. These summaries, including the overall summary, should be coordinated such that the panel proposal is a sensible whole and not a loosely coupled collection of parts. Each member's summary should be approximately 500 words. Please submit six (6) copies of the proposal (cover page, overall summary and member summaries) no later than December 12, 1988 to: IJCAI 89 c/o AAAI 445 Burgess Drive Menlo Park, CA 94025-3496 USA Chairpersons for proposals will be notified of the final decisions by March 27, 1989. The proposals selected for presentation will be published in the proceedings. Chairpersons and members of these panels will be allowed to submit extended versions of their summaries. Revised versions will be due by April 27, 1989. ------------------------------ Date: Wed, 3 Aug 88 13:36:58 EDT From: "Susanne M. HUMPHREY" Subject: dissertation abstracts [Note: Part 5 of 5 - Ed.] .[ AN University Microfilms Order Number ADGD--80234. AU OBEID, NADIM. IN University of Essex (United Kingdom) Ph.D 1987, 113 pages. TI A MODEL OF INFORMATION GROWTH. DE Computer Science. AB Available from UMI in association with The British Library. Requires signed TDF. It is beyond controversy that reasoning about knowledge and belief is a major research issue in Artificial Intelligence (AI). It is also notable that a study of knowledge/belief is included in any AI system irrespective of its aims and motivations. Therefore, for any AI system to have some degree of flexibility, generality and extendability, it must somehow embody some knowledge about how knowledge/belief can be represented, used and possibly revised. In this thesis we introduce a model, called "a model of information growth", which aims at capturing the activity of a rational agent capable of drawing some conclusions from incomplete and/or imprecise information about the domain of concern. The main components of the model consist of: a state representing what the agent accepts at a certain moment, an accessibility relation which determines the set of states which the agent considers possible and a change function which represents its revision policies. In chapter two we introduce the model, discuss its general features and some of the particular choices that are made regarding different concepts such as truth vis-a-vis acceptance and conditionals vis-a-vis acceptance-conditionals. In chapter three we present two nonmonotonic logics that are suitable to reason with incomplete and/or imprecise information. Both logics are proven to be sound and complete with respect to their respective semantics. In chapter four, the non-monotonic logics presented in chapter three are extended with a notion of conditionals in order to represent the agent revision strategies. The resulting logics are proven to be complete and sound. A comparison with Lewis's and Stalnaker's conditionals is made. The model is useful at both theoretical and practical level. Its applications include any branch or area of research in AI that requires reasoning about knowledge and belief. .] .[ AN University Microfilms Order Number ADGDX-80375. AU PENSULO, EMILIUS MWAPE. IN University of Aston in Birmingham (United Kingdom) Ph.D 1987 308 pages. TI INTEGRATING COMPUTER-AIDED ENGINEERING FUNCTIONS: THE MANAGEMENT OF INFORMATION. DE Engineering, System Science. AB Available from UMI in association with The British Library. The integration of computer aided engineering functions is one of the most pressing requirements in the industrial computing field today, as evidenced by the interest shown in the concept of computer integrated manufacturing (CIM). However, major obstacles exist at many levels. Within individual enterprises, the main problem seems to be the inability of discrete software systems to share data effectively. This was the problem addressed in this research work. The solution was seen to lie in a full analysis and centralized management of the common data in the system. There are three main sections in this thesis. Firstly, the formulation of a generalized methodology for the realization of a discrete mechanical product is outlined in order to expose the domain for integration. Specific proposals are then presented regarding the design of integrated CAE systems based on the database approach to information management. These proposals arose from a study of the specific nature of engineering data and activities and also from those of discrete mechanical products. The main problem identified was the maintenance of data integrity in an engineering database: this led to the two main proposals made in the thesis i.e. the introduction of deductive capacities in the conventional database system architecture and the layering of the database structure into a global and several local components. The last section of the thesis concerns the design, implementation and evaluation of an integrated CAE system, created in order to enable an objective evaluation of the proposals arising from the research to be carried out. The specific application chosen was the design of industrial gearboxes. It is now generally recognized that integration has to involve some form of centralized database management. The subject of this thesis is therefore likely to remain of interest for sometime to come. .] .[ AN University Microfilms Order Number ADG88-03441. AU GRICE, ROGER ALAN. IN Rensselaer Polytechnic Institute Ph.D 1987, 438 pages. TI TECHNICAL COMMUNICATION IN THE COMPUTER INDUSTRY: AN INFORMATION-DEVELOPMENT PROCESS TO TRACK, MEASURE, AND ENSURE QUALITY. DE Information Science. AB The needs of the information society have thrust into prominence those who produce the technical information needed to work with computers. As a consequence of society's need for technical information, those who develop and produce that information are called upon to translate the technical capabilities of complex computer systems into a form that enables readers to perform needed tasks. Since the information they produce is an integral part of the product, it follows that these "information developers" must work as an integral part of the product-development team, not as post-facto wordsmiths who clean up the writing done by product developers. To function in this capacity, they need an information-development process that is consistent with the product-development process. They must adapt their writing skills and strategies to the product-development environment in which they work, while at the same time meeting the needs of their audiences, a group that continues to grow in number and diversity as the use of computers evolves. Developing information in this environment is not simply following a set of rules for communicating correctly, but is a process for converting detailed technical information into information that enables users of high-technology products to do the tasks that they want or need to do; it is not merely describing the structure of those products and the functions that are available. Rather, it is a rich, detailed process that involves gathering requirements; setting objectives; producing specifications to meet those objectives; producing drafts according to those specifications; editing, reviewing, and testing those drafts; producing final versions of the information and distributing them to customers; updating and redistributing that information in response to product changes and reader response; and measuring and assessing the quality of the information that has been produced and the process used to produce it. While the process is well defined in many of its aspects, it also possesses many implications for further research and development efforts by academic researchers and industrial practitioners alike. .] .[ AN University Microfilms Order Number ADG88-03073. AU HU, CHENGREN. IN University of Illinois at Urbana-Champaign Ph.D 1987, 277 pages. TI AN EVALUATION OF ONLINE DATABASE SELECTION BY A GATEWAY SYSTEM WITH ARTIFICIAL INTELLIGENCE TECHNIQUES. DE Information Science. AB The tremendous growth of computer-readable databases makes automated database selection very desirable in online services. Gateway systems as components in online networks convert information among online systems and aid the users in various ways including database selection. An evaluation of online database selection by an existing gateway system (InfoMaster, a version of EasyNet) was performed, comparing database selections made by experienced reference librarians with those made by inexperienced searchers aided by InfoMaster for the same set of queries. The results show that automated database selection by InfoMaster still heavily depends on a human decision concerning the broad subject area into which a query falls. When the gateway user properly selects the subject area for a particular query, InfoMaster can do automated database selection as well as the human intermediaries. Artificial intelligence techniques (AI) for automated database selection by gateway systems were studied and a framework of AI techniques for this particular function was outlined based on the analysis of the evaluation results. .] .[ AN University Microfilms Order Number ADG88-03517. AU STERN, RICHARD ELLIOT. IN Rutgers University The State U. of New Jersey (New Brunswick) Ph.D 1987, 129 pages. TI UNCITEDNESS IN THE BIOMEDICAL LITERATURE: AN EXPLORATION OF BIBLIOGRAPHIC CORRELATES. DE Information Science. AB The purpose of the study was to determine if selected bibliographic characteristics of biomedical journal papers are associated with the subsequent uncitedness of the papers. Two characteristics specific to the biomedical literature and seven general characteristics were examined. The biomedical characteristics include whether or not the research was funded by the National Institutes of Health and the biomedical research level which is a four level classification of journals according to their research orientation: Level 1, Clinical Observation; Level 2, Clinical Mix, Level 3, Clinical Investigation, and Level 4, Basic Research. The general characteristics include the number of authors, title words, key title words, and references; also, the age, price, and circulation of the journal in which the paper was published. The rate of uncitedness, using the 275 source journal set as the citing set is 13.3%. The rate of uncitedness using the Science Citation Index as the citing set is estimated to be 4.4%. Research support and biomedical research level are significantly associated with uncitedness. Uncitedness is lower for NIH-supported papers. Uncitedness decreases with biomedical research level, with uncitedness lowest among papers published in basic research journals. A discriminant analysis function with the general bibliographic characteristics correctly predicted cited and uncited papers with 70% accuracy. Uncited papers alone were predicted with 75% accuracy. In partitioning the components of the discriminant model, the number of references accounted for 76.4% of variability, and the number of authors accounted for an additional 9.7% of the variability. This reinforces the finding that the number of references and authors are significantly related to uncitedness. Uncitedness papers have fewer references and authors compared to cited papers. Future studies of uncitedness should proceed in two directions. One, additional bibliographic characteristics such as author affiliation and length of paper ought to be tested. Second, further study of the significantly associated characteristics ought to be examined more closely to determine whether their numbers differentiate between levels of quality in papers or whether there is some other mechanism at work that accounts for the significant differences in rates of uncitedness. .] .[ AN University Microfilms Order Number ADG88-03222. AU THORNBURG, GAIL ELLEN. IN University of Illinois at Urbana-Champaign Ph.D 1987, 111 pages. TI LOOK: IMPLEMENTATION OF AN EXPERT SYSTEM IN INFORMATION RETRIEVAL FOR DATABASE SELECTION. DE Information Science. AB This project was designed to construct an advisory or expert system in one area of online information retrieval, specifically, choice of online database. Problems of this domain include proliferation of potentially expensive databases, and the difficulty of predicting the specific database(s) most likely to yield optimum results in the climate of extreme "information compartmentalization." This implementation made use of an integrated set of software tools developed in the Artificial Intelligence Lab of the Department of Computer Science at the University of Illinois at Urbana-Champaign. The system was designed to reflect as closely as possible the decision-making expertise of academic online searchers in life sciences. The implemented system, LOOK (non-acronymic), represented general features of 18 online databases, and its advice succeeded in satisfying the experts involved in its development. The system used a rule-based representation, and advisory sessions were guided by an inference algorithm featuring three phases of evaluation. Issues discussed here include the numbers of variables and values required to represent the domain with any adequacy, the levels of abstraction apparent in these variables, and the difficulty of separating domain from world knowledge in constructing an apt representation. .] .[ AN University Microfilms Order Number ADG88-03480. AU GOODIN, M. ELSPETH. IN Rutgers University The State U. of New Jersey (New Brunswick) Ph.D 1987, 186 pages. TI THE TRANSFERABILITY OF LIBRARY RESEARCH SKILLS FROM HIGH SCHOOL TO COLLEGE. DE Library Science. AB The specific purpose of this research was to investigate three interrelated questions: First, can a program of instruction be developed to teach high school seniors efficient and effective search strategies and information-gathering skills useful to them as college freshmen? Second, can these skills be reflected in a research paper? Third, how can these skills be measured for transferability? A program of instruction was developed and taught by the high school librarian that included the responses and recommendations of 62 college faculty members to a questionnaire on library-related course requirements. Participants in the study, 159 seniors from the college preparatory English classes of two comparable high schools, were divided into one control and one experimental group for a pilot test (Study I) and for a second test (Study II). Students in the experimental groups were given a pre-test on basic college library information knowledge, followed by a series of lessons on the research process, the assignment of a research paper, a post-test on basic college Library information knowledge, and the return of their graded papers, the grading being completed by two college English professors. Students in the control groups were given the pre-test, the research paper assignment, the post-test and the return of the graded papers. A Likert-type attitude scale questionnaire was administered to all high school students, and to the participants in Study I during their first semester in college. High school students receiving the instruction scored significantly higher on the post-test than did students not receiveing the instruction. These skills were reflected in a research paper acceptable at the college freshmen level. College students exposed to the program of instruction indicated they were able to effectively utilize the research skills learned in high school when conducting research to meet course requirements. There were no significant differences between the groups on the attitude scale questionnaire, indicating an inability of this questionnaire to adequately measure the transferability of research skills from high school to college. Finally, the high school librarian became a linking agent between the high school and college libraries, by making the librarian in the research setting approachable. .] .[ AN University Microfilms Order Number ADG88-03160. AU OLDEN, EDWARD ANTHONY. IN University of Illinois at Urbana-Champaign Ph.D 1987, 175 pages. TI THE BENEFICIARIES OF LIBRARY AND INFORMATION POLICY IN BRITISH AND EX-BRITISH AFRICA: STEPS FROM THE WHITE WOMEN'S LEAGUE TO THE ELECTRONIC LIBRARY. DE Library Science. AB This dissertation examines the beginnings and early years of five libraries set up with the stated aim of serving the public in British and ex-British Africa, and argues that the beneficiaries of library and information policy have not been the public at all but only a small minority: the elite and the would-be elite. It suggests a pattern from the 1920s and 1930s to the present: from the fee-charging whites-only leisure reading libraries of the settler colonies to the computerized information services recommended for the use of the black "policy-makers" and "decision-makers" of today. Five episodes from Kenya, the Gold Coast, and Nigeria are used to support the argument: the fee-charging whites-only service that the Carnegie Corporation of New York--the champion of the free public library--sponsored in Kenya Colony in the 1930s; the fee-charging service for the white and black educated elite of Lagos, Nigeria, that Carnegie money helped bring into being around the same time; the fee-charging service that the British Council introduced for the white and black educated elite of the Gold Coast and Nigeria in the mid-1940s; the Northern Nigeria Regional Library, set up in 1952, which with time concentrated more and more on the capital and less and less on the rest of the region; and the National Library of Nigeria, established with Ford Foundation help in the 1960s to assist in the "responsible conduct of modern government." Accounts are based on Carnegie, Ford, and British Council archives, and on published sources. South African, Rhodesian, and other examples are also mentioned. The dissertation attempts to show that librarianship and information science are very much part of the political context in which they operate; that only a small percentage of the African population has benefitted from the western-style libraries and information services that have been introduced; and that the study of such matters adds one more dimension to what is known about cultural and other forms of dependency. .] ------------------------------ END OF IRList Digest ********************