共查询到20条相似文献，搜索用时 31 毫秒
1.
《图书馆管理杂志》2013,53(1-2):191-193
Abstract This work explores the possibility of using unrestricted machine-readable full text to augment the indexing vocabulary of the Dewey Decimal Classification and make it more accessible to end users. It is based on the hypothesis that contemporary text is more likely to contain up-to-date vocabulary than a reference work that must undergo a rigorous editorial review process. 相似文献
2.
《Journal of Business & Finance Librarianship》2013,18(4):35-36
Abstract This study examines the extent to which full-text business databases support scholarly research in finance. A list of articles cited by finance researchers was compared to the title lists of three full-text business databases: ABI/INFORM (ProQuest Information and Learning), Business Source Premier (EBSCO Information Services), and General BusinessFile (Gale Group). The most comprehensive coverage was found to be offered by ABI/INFORM, which provided indexing for 76.0% of the articles and full text for 35.9%. The articles shown as full text in the title lists were then searched in the databases to provide a measure of reliability of the title lists. 相似文献
3.
《图书馆管理杂志》2013,53(1-2):41-48
Abstract The primary goal of the ELIXIR project was to achieve improvements in the performance of text-retrieval systems by confronting some of the challenges presented by the indexing of full-text documents. We explored two hypotheses involving the use of linguistic analysis to improve the automatic creation of indexes: (1) if it is possible to develop algorithms that index only the words which most closely capture the document's content-not the complete text, as is done in existing sys-tems-the retrieval of irrelevant documents might be minimized; (2) if it is possible to create indexing algorithms that perform a minimal syntactic parse required to capture important phrases, the index may represent the document's content even more accurately, increasing the likelihood that the article can be retrieved by a relevant query. 相似文献
4.
5.
《Behavioral & Social Sciences Librarian》2013,32(1):17-25
ABSTRACT The article examines the weaknesses and strengths of the economics database, EconLit. Comparison is made with other social science databases with especial emphasis on the alphanumeric classification and indexing system unique to EconLit. The general conclusion is that, despite the high value of its information content, EconLit would benefit by adhering closer to the indexing practices of other mainline social science databases. 相似文献
6.
Don C. Smith 《广播与电子媒介杂志》2013,57(1):35-44
Difficulties in indexing the degree of motivation or attention to television by people in the same room with a “live” television set have long been recognized. One element of this problem has been the question of determining which member of the household actually makes the selection of specific program or channel, and whether there are differences between types of households or socio‐economic classes that are related to different patterns of channel or program selection. 相似文献
7.
McDonald S 《Health information and libraries journal》2002,19(1):14-20
Introduction: Locating reports of trials from journals not indexed in the major databases presents difficulties to systematic reviewers, and may be a factor in improving the reliability of the reviews. Objectives: To identify and make available reports of controlled trials from the Australasian Medical Index (AMI). To measure the quality of indexing of trials in AMI. Methods: Using a highly sensitive search strategy consisting of methodology indexing and free‐text terms, records from AMI were read for reports of controlled trials. Trials meeting the criteria were submitted for inclusion in The Cochrane Controlled Trials Register (CCTR) and assessed for the quality of their indexing. Results: 3621 records were downloaded, of which 512 were identified as reports of controlled trials (317 RCTs; 195 CCTs) and submitted to CCTR. The precision of methodology indexing terms was 60%, but sensitivity just 18%. The quality of indexing of trials was generally poor with only 50 tagged with the RCT/CCT publication type term. 453 reports (88%) were not previously available in CCTR. Conclusions: The large proportion of trials found to be unique to the AMI database increases the pool of studies available to systematic reviewers, and helps ensure CCTR remains the most comprehensive source of trials. 相似文献
8.
《Behavioral & Social Sciences Librarian》2013,32(2):119-122
Abstract There are a wide variety of Web-based periodicals available free of charge, including e-zines and scholarly journals. This column describes several useful Web sites that provide access to and indexing for online journals in the behavioral and social sciences. 相似文献
9.
关键词自动标引是一种识别有意义且具有代表性片段或词汇的自动化技术。关键词自动标引可以为自动摘要、自动分类、自动聚类、机器翻译等应用提供辅助作用。本文利用基于知网的词语语义相关度算法对词汇链的构建算法进行了改进,并结合词频和词的位置等统计信息,进行关键词的自动标引。实验证明,该方法可以有效的进行关键词的自动标引。 相似文献
10.
11.
提出一种基于潜在语义索引和本体论的文本语义处理方法。首先构建一个基于本体论的虚拟标准文本特征向量，然后采用潜在语义索引方法以虚拟标准文本特征向量为参照对文本集进行语义聚类，最后在虚拟标准文本特征向量的导引下利用本体库中的知识对聚类获得的文本集合的类别和语义进行显性标注。实验表明，该方法能较好地在语义层面对文本进行有效的聚类，而且聚类结果能显性地显示类聚所属的类别。 相似文献
12.
一种改进的余弦向量度量法文本检索模型 总被引：2，自引：1，他引：1
13.
《期刊图书馆员》2013,64(3-4):287-297
Summary ScienceDirect? has been developed by Elsevier Science to provide a databased, Web host environment for its journals and those of other participating publishers. The target is to create an integrated environment, which will permit efficient and transparent access to the scientific literature by integrating abstracting and indexing searching and by linking to other publishers' sites. 相似文献
14.
Filiberto Felipe Martínez-Arellano Federico Hernández-Pacheco Efrén Chávez-Hernández 《Cataloging & classification quarterly》2013,51(2-3):91-105
AbstractSubject indexing and classification of law resources is a complex issue due to several factors: specialized meanings of legal terms, meanings across different branches of law, terms in legal systems from diverse countries, and terms in different languages. These issues led to the development of a classification and subject indexing system which will help answer the major challenges of indexing and classifying law resources in the Research Institute Library at the National Autonomous University of Mexico. Adopting its own classification required interdisciplinary work between law and information organization specialists, constant updating by legal specialists and others beyond the Legal Research Institute; and the sharing of this classification system with other institutions. Now, this classification system is used by important institutions that specialize in law, such as the network of Libraries of the Supreme Court of Justice of the Nation of Mexico. The purpose of this article is to show why and how this law classification and subject system was developed and is continuously being updated by libarians and law scholars in order for it to meet their specific needs. 相似文献
15.
This article describes the process of transforming the Arkansas Periodical Index, which began as a card file index in the University of Arkansas (U. of A.) Libraries, to an online database. As in many other states, in-house indexes were compiled to cover local history journals and subject-related magazines. A variety of software now exists to mount locally created databases, one option being the choice to use the same turnkey software that runs the library's online catalog. We began by converting our card files to a local Inmagic DB/Textworks database. The U. of A. received the 1991 Gale Research Award for Excellence in Reference and Adult Library Services for that design. We later mapped the database structure to MARC fields and mounted it as a Web-accessible database through our library OPAC vendor, Innovative Interfaces, Inc. (III). The authors begin with a brief history of periodical indexing, the reasons for choosing MARC formatting, and the field mapping required. The article explains the development of the database to include other publication formats, such as material from biographical compendiums, newspaper indexing, and university publications. The mapping incorporates indexing several publication formats in the same database. It also provides fields for linking citations back to the OPAC records, where users may find call numbers for journals or books, be connected to the full text of articles when available, or to an association's own website when applicable. We address some of the concerns for libraries in developing indexing projects. The article offers a mapping table of migrated fields that can be used by others wanting to encourage the rediscovery of local print periodicals and to expand access to the various materials otherwise buried in the library's special collections. The Arkansas Publications Index was released in April 2005 and is available free at http://arkindex.uark.edu/. 相似文献
16.
17.
18.
研究构建了具有位置信息控制的特义禁用词语义环境，进而运用于中文文献元数据CXMARC文本的自动标引和主题信息的数据挖掘，其中研究设计的预处理特义中文禁用字词切分算法SWF，能有效地减少领域的分词歧义性和缩短标引时间，从而改进了传统最大匹配MM算法的自动标引质量和效率。 相似文献
19.
20.
文本检索的潜在语义索引法初探 总被引：5，自引：0，他引：5
传统的文本检索方式是基于提问集合和文本集合的单纯语词匹配检索,然而这并不能解决检索实践过程中存在的同义和多义问题。文章阐述了文本检索的潜在语义索引法的原理并通过实验来验证潜在语义索引可以用来解决同义和多义问题,完善检索系统的性能。 相似文献