知能型 情報檢索 시스팀에 關한 考察

李英子 慶北大學校 大學院 文獻情報學科 同窓會 1989 圖書館·情報學硏究 Vol.1 No.-

Some conclusions derived form the study are as follows : ① The systematic of cognitive science explaining the cognitive structure of individuals as information sources, information retrieval systems, and searchers should be the basis of the design and understanding of the intelligent information retrieval systems. ② Due to the various reasons, information retrieval can't be an ideal area for the application of artificial intellligence techniques yet. ③ The influence of the expert system has the research on the information retrieval to shift from the paradigm concerned with retrieval algorithems to the one in which users, retrieval heuristics, knowledge, human-computer interaction can be the key themes. ④ Many experimental researches as well as a few operational prototypes for the intelligent information retrieval systems provide some evidence for the quality enhancement in the retrieval processing by meance of expert systems, knowledge representation and natural language processing. ⑤ For the refinement of the present intelligent information retrieval system, it should be necessary that information specialists pay attention to the solution of the problems on the lack of real-world knolwedge which can be the critical bottleneck for the devleopment of the systems.

도서관에서 LAN시스팀의 적용

이강욱 慶北大學校 大學院 文獻情報學科 同窓會 1992 圖書館·情報學硏究 Vol.2 No.-

In the aspect of library automation, this paper, describes the outline of LAN technique. The purpose of LAN is to link computers and peripherals, allowing multiple users to run individual applications and also to access the network to transfer data and to share programs and peripherals linked to the LAN. Among the topologies, Star, Ring, Bus, the bus network may be most sufficient to academic library, because of its advantages(extensibility, efficiency in operation‥‥‥et al.). The measure of maximum potential utilization, results are derived as follows : 1) The smaller the mean frame length, the greater the difference in maximum mean throughput rate between token passing and CSMA/CD. 2) Token passing is the least sensitive to work load. 3) CSMA/CD offers the shortest delay under light load, while it is most sensitive under heavy load to the work load. In the library market specifically, it is safe to forecast that baseband system utilizing coaxial or twisted pair wire will dominate the local area network product offerings to the future. As videotext and the electronic library evolves, broadband networks capable supporting voice and video signals as sell as written data may begin to appear. There is no doubt that interest in and use of microcomputer LANs by library is growing and will continue to grow. Regardless of timing, local area network will help libraries to reduce the cost of automation and enhance patron services.

公共貸出權 制度에 관한 硏究

柳仁順 慶北大學校 大學院 文獻情報學科 同窓會 1992 圖書館·情報學硏究 Vol.2 No.-

Public Lending Right is the idea that an auther is entitled to be compensated for the multiple uses of his copyrighted books in libraries. While some form of a public lending right is already a legal fact in twelve nations, this has come about only over the past thirty-five years and actual discussion of the principle cannot be said to have begun in earnest until shortly before 1920. Two arguments in the PLR can be summarized as follows : 1. Most author's view points for PLR are based on three points : ① that the author's proprietary rights to the texts they have written are fixed and inalienable natural rights, ② that these rights are unfairly in fringed upon by libraries which freely circulate the books embodying these texts, and ③ that the effect of such infringements materially deprives the authors of sums they would otherwise realize through private purchase of these books. 2. Most librarian's viewpoints for PLR are based on three points : ① that the application of PLR on libraries become financial and administrative burdens, ② that the PLR is to infringe for the idea of a free public library, ③ that the idea of PLR is to band the right of individuals to buy books and circulate these books freely. Nevertheless, Nations that established to PLR are increasing gradually. The application of PLR involves making payments to authors which are related in someway to the use of their books in libraries. Since large numbers of authors, books, loans, and libraries are involved, there is no simple way each library can be made directly responsible for transmitting money to authors, whatever funding system may be adopted. The major problms on the application of PLR in Korea are, as it were, do authors actually suffer financial damage through library circuation and what are the effects of PLR on libraries.The result of this study, the time is not yet ripe for the application of PLR in Korea.

「根源에서 思惟하는 讀書」

李千孝 慶北大學校 大學院 文獻情報學科 同窓會 1992 圖書館·情報學硏究 Vol.2 No.-

電子圖書館에 관한 硏究

金正賢 慶北大學校 大學院 文獻情報學科 同窓會 1989 圖書館·情報學硏究 Vol.1 No.-

This study is an attempt to furnish some helpful data for the design and implementation of electronic library system based on the analysis of cases of electronic library model. To begin with, the concepts, historical developement, functions and electronic media were overviewed in the second chapter, on the basis of which the concrete cases were introduced in the third chapter ; Dowlin's model knowed by the most ideal model of electronic library, PPLD's(Pikes Peak Library District) Maggie Ⅲ system, Doong-Ji's ELIS(Electronic Library Information System), the model of EL Study Group and etc. In the future by constructing the electronic library system, library services will be extended to the neographic age, bibliotopia and global village library.

國際十進分類法의 起源과 發展過程

李昌洙 慶北大學校 大學院 文獻情報學科 同窓會 1989 圖書館·情報學硏究 Vol.1 No.-

The purpose of this study is to find out the original idea and development of Universal Decimal Classification. The results of the study can be summarized as follows : 1. In 1895, following the first International Conference on Bibliography, the newly-founded Institut International de Bibliographie (IIB) sponsored a scheme initiated by two Belgians Paul Otlet and Henri La fontaine. The scheme was prepared for the classification of all published literature, so that a comprehensive classified index to all literature could be produced. 2. By an arrangement with Melvil Dewey, who had just published the 5th edition(1894) of Dewey Decimal Classification(DDC), the IIB began expanding the schedules part by part. UDC was originally issued in 35 parts, from 1899 to 1905. In 1905 was published the first complete international edition in French. Thus first complete international edition was orginally derived form the DDC (5th, 1894). 3. UDC editions are published full editions,abridged editions, medium editions and special subject editions in least 20 languages. 4. IIB changed into name to Federation Internationale de Documentation(FID) in 1937. FID is the body which has ovrall responsibility for UDC. It works through its international committee on universal classification. National member committees are represented on it. Day-to-day control lies with the Central Classification Committee, called FID/CCC. Each member nation is supposed to have a national committee, which is made responsible for the editions in that language. Besides these administrative committees, there are international and national subject committees. The international committee reports to FID/CCC. Similarly, the national subject committees report to the concerned national committee. 5. UDC can be used for the classification of articles and monographs. It may also be used for classifying items to be included in catalogues, indexes and bibliographies. UDC is used by a large number of libraries around the world. It has been popular in Enrope, Latin America and Japan. In the USSR and the countries of East Europe, its adoption was mandatory. UDC is now the official classification scheme in scientific and technical libraries in USSR. UDC is used by a number of abstracting and indexing services, either as the basis of their arrangement, or else as an additional aid to users. UDC class numbers appear on MARC tapes.

大學圖書館 貸出業務 改善에 關한 小考

李相根 慶北大學校 大學院 文獻情報學科 同窓會 1992 圖書館·情報學硏究 Vol.2 No.-

The aim of this study is to grasp the existing problems in circulation work of university library and to understand the desirable improving direction about this. The results of the study can be summarized as follows : 1. University library has to introduce the bar code system as a data input apparatus in book loans and return so as to treat a large amount of book loans and return work quickly and exactly. 2. University library should consider the formation of loans master file, if possible, after his data base is constructed, and after retrieval function is completed. 3. Through loans enquiry about specific materials (confirming about borrower of materials) and reservation of book, library should elevate the efficiency of use of materials. 4. Through a report about prearrangement of book return and a report about delay (overdue), library should attain the efficiency of book management and diminish damage and loss of book, and prevent a phenomenon of circulation stagnation. 5. All statistical data generated from loans and return work become automatic production, then, times and efforts will be economical. 6. Applying the bar code system, the library can be carry out inventory as scanning the book number in bar code stuck to book by a portable scanner. As mentioned above, when the improvement of work is accomplished, university library has also the function of furnishing information in addition to an usual function of offering books through loans, then, in the end, can be bring such effect as enlargement of service about users.

인공지능에서의 지식표현에 관한 고찰

김희섭,우동진 慶北大學校 大學院 文獻情報學科 同窓會 1992 圖書館·情報學硏究 Vol.2 No.-

The symbiotic integration of, on the one hand, what might be termed traditional, and on the other hand, AI-derived computing techniques is likely to increase. We are moving away from a situation in which the novelty and distinctive historical origins of AI(Artificial Intelligence) techniques have caused them to be often perceived, and applied, separately from more traditional mainstream approaches. The Knowledge Representation, Inference, Heuristics and AI Programming Language are basic components of AI. Among these components, the Knowledge Representation is currently the most emphasized area in the field of AI. And the performance of AI system depends on the quality and the quantity of the knowledge. This paper attempts to introduce various techniques of the Knowledge Representation and to espound each method, its advantages and disadvantages, examples of each application system, etc. Chapter I of this paper contains introductory remarks. The of the Knowledge Representation is introduced in Chapter Ⅱ. Chapter Ⅲ Expounds Logic methods, Semantic Networks, Conceptual Dependency, Frames, Scripts and Production Systems, Chapter Ⅳ explains the Procedual methods of knowledge representation. Rather than thinking of complex problems and trying to devise techniques to solve them, it is often more productive to identify problems analogous to those to which techniques have been successfully applied.

정보검색 시스팀에 Fuzzy Set이론의 적용

李淳在 慶北大學校 大學院 文獻情報學科 同窓會 1989 圖書館·情報學硏究 Vol.1 No.-

An automatic information retrieval system should provide adequate ffacilitices for expedient transfer of knowledge from document to users of these documents who request it. Because of a very large volume of sources, the transfer has to be selective, the selection criteion being specified by the user requesting the information. Although some commercially available systems are now widely used, a great deal of difficult practical and theoretical problems in his field remain to be solved. In most operational information retrieval systems the standard retrieval methods based on set theory and binary logic are used. These methods would be much more attractive if they could be extended to include the importance of various index terms in document representations and search request formulations, in addition to a weighing mechanism which could be applied to rank the retrieved documents. This observation has been widely recognized in the literature as such extended retrieval methods could provide the precision of a Boolean search and the advantages of a ranked output. However, a closer examination of all the reported work reveais that up to the present the only possible approach of sufficient consitency and rigorousness is that based on recently developed fuzzy theory and fuzzy logic. As the concept of a fuzzy set is a generalization of the conventienal notion of a set, the generalization of the information retrieval methods based on set theory and binary logic can be derived in a natural way. This paper presents and evaluation of some of the existiong techniques of document retrieval, examining their weak points. Then it outlines new techniques based on fuzzy sets and logic. The advantages of these new tochniques are discussed and a method of practical implementation of a fuzzy information retrieval system is briefly outlined.

敎育學部門의 索引言語 階層性 比較分析 : 한국십진분류표와 KEDI교육 THESAURUS를 중심으로 Centering on the KDC and KEDI EDUCATIONAL THESAURUS

裵永活 慶北大學校 大學院 文獻情報學科 同窓會 1989 圖書館·情報學硏究 Vol.1 No.-

The purpose of this is to understand the hierarchy of concepts between KDC and KEDI EDUCATIONAL THESAURUS which are used terms in education. The results of the study can be summarized as follows : 1) The number of the terms of agreeing to the relation of hierarchy is few in the index language of KDC and KEDI EDUCATIONAL THESAURUS. A large number of the terms of agreeing to the relation of hierarchy don't help greatly to the indexing and document retrieval. 2) Terms of disagreeing to the relation of hierarchy are made to express the difference in quality of concepts and extents of cognition or characteristics of classification and thesaurus, for example professional education, curriculum etc. 3) Among the ambiguous terms to the relation of hierarchy, the number of terms equaling to the position of hierarchy is a few and the number of terms unequaling to the position of hierarchy is many. Therfore, it shows that guidance to the more special item is insufficient in the specific subject. Finally, as the descriptors of KEDI EDUCATIONAL THESAURUS show the irrationality in the concepts and adequacy in the relation of hierarchy with the cultural and cognitional difference and difficulty of vocabulary selection in translation, it would be desirable that descriptors should be regulated to the rationality and adequacy.