• Journal of the Korean Data and Information Science Society
• /
• v.24 no.5
• /
• pp.999-1011
• /
• 2013

In this research, we propose a document recommendation method which can find documents that are relatively important to a specific document based on citation information. The key idea is parameter tuning in the Neumann kernal which is an intermediate between a measure of importance (HITS) and of relatedness (co-citation). Our method properly selects the tuning parameter ${\gamma}$ in the Neumann kernal minimizing the prediction error in future citation. We also discuss some comutational issues needed for analysing large citation data. Finally, results of analyzing patents data from the US Patent Office are given.

• Journal of the Korean Society for information Management
• /
• v.23 no.1 s.59
• /
• pp.181-200
• /
• 2006

The purpose of this study is to analyze the accessibility of web resources contained in the references section of journal articles.1 was identified a total of 1,377 web citations for a major journal in information science over a ten year period. The results show that the accessible ratio of domestic web resources was at 40% and the ratio of foreign web resources about 50%. The accessible ratio of domain type web resources was shown 58% while the same ratio for file type web resources was 45%. This low accessibility of we resources in references poses threat to the overall value of journal articles.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.21 no.1
• /
• pp.211-221
• /
• 2010

The purpose of this study is to verify if there is any difference on researchers' language preference for citing documents in different subject fields. 8 scientific journals were selected one of each from 8 main categories, given by National Research Foundation of Korea. Five variables including the rate of citing domestic and foreign documents for each journals were choose, and the differences on language preference among researchers of different subject fields were checked by implementing ANOVA. As a result, there are some differences on language preference, in terms of the average number and percentage of foreign documents, among researchers, based on their subject background. It also found that 8 subject categories divided into 4 small clusters.

• Journal of the Korean Society for information Management
• /
• v.16 no.2
• /
• pp.7-26
• /
• 1999

This experimental study examines informetric analysis of World Wide Web based upon cocitation analysis of Web pages and features of Web resources in the field of communication studies. Cocitation analysis is basically performed to examine the intellectual structure of the communication studies in reflecting link count on the Web. The selected Web resources in the field are mapped in two dimensions based upon the similarities of cocitation frequency, correlation matrix, mutidimensional scale and cluster analysis. Cocitation analysis methods using organizational homepage, personal homepage, or Web index, to Web produced clustering of Web resources that had topical similarities. So far, although informetric analysis of Web resources is in the preliminary stage, it shows that Web can be a new tool for indicating the intellectual structure of a specific research field. In addition, this study analyzes characteristics of printing resources and Web resources, and differences of research methods in applying cocitation analysis.

• Journal of the Korean Society for Library and Information Science
• /
• v.37 no.1
• /
• pp.61-79
• /
• 2003

In order to clarify the trend of study by the researchers in the field of library and information science. This study is to analyze the thesis of library and information science researchers in various aspects which is included in LISA and ISA, the representative database in library and information science. In addition, the study is to analyze the form of citation of domestic library and information science researchers by studying the citation written in the thesis of the database according to the classification of form and nationality. The study made a conclusion that the journal is most frequent]y cited regarding forms, which is expected to promote the internationalization of library and information science and indicate the direction of our study in the future.

• Journal of the Korean Society for Library and Information Science
• /
• v.37 no.1
• /
• pp.247-268
• /
• 2003

Recently, the way information is accessed has changed significantly because of the rapid growth of the Internet and a move toward electronic Publishing. With that condition, there in be a need to efficiently determine and extract relevant information from digital documents. This study has proposed that template mining can be used for extracting different kinds of information from digital documents, and described research of reference ]inking system and then built temp]ate for citation information using analysis result of sample articles. Also, this study has shown the potential of template mining to automatically create citation databases.

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.174-182
• /
• 2004

In order to produce information the author usually reference other authors' work. A citation index leads users to papers by citations. Citations lead the user to desired information. In this paper, KSCI(Korea Science Citation Index) which defines the relationships between citing documents and cited documents has been constructed. KSCI System is to solve problems for recursive retrieval in ISI's SCI(Science Citation Index) Path Encoding Indexing technique was used to solve the problems. From the analysis of data, this system has efficiency about 8.98% in the aspect of data storage. In the aspect of retrieval, there was efficiency between citing documents and cited documents, especially there was over 40% of efficiency in the retrieval of cited documents. It is concluded that suggested KSCI system will provide efficient storage and retrieval system.

• Journal of the Korean Society for information Management
• /
• v.27 no.2
• /
• pp.241-252
• /
• 2010

Titles have been regarded as having effective clustering features, but they sometimes fail to represent the topic of a document and result in poorly generated document clusters. This study aims to improve the performance of document clustering with titles by suggesting titles in the citation bibliography as a clustering feature. Titles of original literature, titles in the citation bibliography, and an aggregation of both titles were adapted to measure the performance of clustering. Each feature was combined with three hierarchical clustering methods, within group average linkage, complete linkage, and Ward's method in the clustering experiment. The best practice case of this experiment was clustering document with features from both titles by within-groups average method.

• Journal of Information Management
• /
• v.24 no.4
• /
• pp.1-23
• /
• 1993

The purpose of this study is to investigate and analyze the obsolescence of documents according to their dates, types and locations of publication. The main results are as follows : The half-life of monographs is 12.09 years while those of articles and research papers are 9.68 and 8.93 years, respectively. Moreover, documents are most often cited by researchers within two years since their publication. Lastly, but not the least, the estimated weeding points for monographs, articles, and research papers, assuming that their weeding points have been realized when their accumulated citation rates reach 90%, are 40.15, 32.15, and 29.65 years, respectively.

• Journal of the Korean Society for information Management
• /
• v.34 no.3
• /
• pp.109-124
• /
• 2017

Through a wide variety of disciplines, practices on data access and re-use have been increased recently. In fact, there has been an emerging phenomenon that researchers tend to use the data sets produced by other researchers and give scholarly credit as citation. With respect to this practice, in 2012, Thomson Reuters launched Data Citation Index (DCI). With the DCI, citation to research data published by researchers are collected and analyzed in a similar way for citation to journal articles. The purpose of this study is to identify the characteristics and intellectual structure of sociology field based on research data, which is one of actively data-citing fields. To accomplish this purpose, two data sets were collected and analyzed. First, from DCI, a total of 8,365 data were collected in the field of sociology. Second, a total of 12,132 data were collected from Web of Science with a topic search with 'Sociology'. As a result of the co-word analysis of author provided-keywords for both data sets, the intellectual structure of research data-based sociology was composed of two areas and 15 clusters and that of article-based sociology was composed with three areas and 17 clusters. More importantly, medical science area was found to be actively studied in research data-based sociology and public health and psychology are identified to be central areas from data citation.