Please use this identifier to cite or link to this item: http://rportal.lib.ntnu.edu.tw:80/handle/77345300/15146
Title: 引用文獻分析為基礎的多評準期刊評選模式之研究
Other Titles: Multi-criteria Approach to Journal Collection Evaluation Based on Citation Analysis Method
Authors: 黃明居
梁詩婉
Ming-Jiu Hwang
Shih-Wan Liang
Issue Date: Oct-2012
Publisher: 國立台灣師範大學圖書資訊學研究所
Graduate institute of library and information studies ,NTNU
Abstract: 大學圖書館每年都需面臨期刊增刪的問題,尤其西文期刊價格昂貴,如何有效追蹤與分析所採購期刊之使用情形供決策者參考,為圖書館重要之課題。本研究以引用文獻分析為基礎,提出機構讀者使用為導向的多評準期刊評選模式,考慮機構讀者引用數量、國內專家所提出核心期刊清單以及國際JCR 所蒐集的高影響係數期刊等三項準則,並以多評準決策的權重法得出最後的期刊購買清單。除了建立評選模式外,本研究所建置的決策支援系統不但具備引文資料分析功能且提供適度評分彈性供決策者調整使用。最後以某研究型大學資工系5 年資料作實證研究,結果顯示本研究之模式與開發的系統,不但可以得出布萊德福分區之期刊資料與引用比率外,亦可清楚看出機構讀者未引用的期刊資料,可作為刪刊之參考。
The decision-makers of academic library need to face the serious problem about the journal collection evaluation every year, especially in the case of in lack of library funding. This paper proposed a multiple criteria evaluation model based on the citation analysis method. Three criteria of this model are (1) the cited number of library patrons; (2) core journal list proposed by Taiwanese experts and (3) JCR journal list of the high impact factor. In addition to the establishment of the evaluation model, a decision support system, of which not only with the citation data analysis but also having a flexible mechanism for decision makers is provided. The results clearly showed that this evaluation model and decision support system can be drawn from the data of cited ratio and the journal list of Bradford's Law Area. It also showed that the institutional patrons never been cited data, this information can be used as important reference for library manager to delete journal title every year.
URI: http://rportal.lib.ntnu.edu.tw//handle/77345300/15146
Other Identifiers: 2A040F71-9BE7-F00D-EE33-DC6040A2EE6B
Appears in Collections:圖書館學與資訊科學

Files in This Item:
File SizeFormat 
ntnulib_ja_A1021_3802_052.pdf1.42 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.