• Journal of the Korean Society for information Management
• /
• v.29 no.1
• /
• pp.157-174
• /
• 2012

This paper compares two recent methods for exploring a scientist's research history: citation identity and self-citation network. The former is proposed by White(2000), while the latter is suggested by Hellsten et al.(2007). An experimental citation analysis was carried out on the research output of Young Mee Chung, a renouned Korean information scientist. The result shows that the two methods divided the research period into two sub-periods in the same way. They also identified the major research themes very similarly. In the analysis of each method's performance in depth, the two methods revealed different functions to understand a researcher's history. Citation identity was useful to identify authors who have affected Chung's research in terms of research topics. whereas, self-citation network was successful to identify the core papers and leading papers of the research sub-periods. This study indicates the combination of two methods can provide rich information on a scientist's research history.

• Journal of the Korean Society for information Management
• /
• v.29 no.1
• /
• pp.115-134
• /
• 2012

The author co-citation analysis is generally based on the frequency of the first author because most citation databases include only the first author in the bibliographic information. In this sense, the purpose of this study is to provide a better knowledge structure by utilizing the multiple authorship of author co-citation analysis. To achieve the purpose of this study, four different data sets are prepared: (1) counting the first author, (2) counting all the author without limiting the total frequency, (3) counting all the author with limiting the total frequency, and (4) counting adjusted frequencies based on the order of author subscription. The findings of this study show that there are clear differences between the knowledge structure counting all the author and the one counting only the first author. In addition, depending on the different methods, there are subtle changes of cluster members for authors.

• Journal of the Korean Society for information Management
• /
• v.25 no.4
• /
• pp.283-308
• /
• 2008

In this study, we testify that network structural attributes of a citation network can explain other aspects of journal citation behaviors and the importances of journals. And we also testify various citation impact indicators of journals including JIF and h-index to verify the difference among them especially focused on their ability to explain an institution's local features of citation behaviors. An institutional citation network is derived using the articles published in 2006-2007 by biotechnology faculties of Y University. And various journal citation impact indicators including JIF, SJR, h-index, EigenFactor, JII are gathered from different service sites such as Web of Science, SCImago, EigenFactor.com, Journal-Ranking.com. As a results, we can explain the institution's 5 research domains with inter-citation network. And we find that the co-citation network structural features can show explanations on the patterns of institutional journal citation behavior different from the simple cited frequency of the institution or patterns based on general citation indicators. Also We find that journal ranks with various citation indicators have differences and it implies that total-based indices, average-based indices, and hybrid index(h-index) explain different aspects of journal citation pattern. We also reveal that the coverage of citation DB doesn't be a matter in the journal ranking. Analyzing the citation networks derived from an institution's research outputs can be a useful and effective method in developing several library services.