標題: 複雜網路應用於圖書館讀者借閱之分析
The Analysis of Library Circulation Record by Complex Network
作者: 李芳讚
Fang-Tsan Li
黃明居
Ming-Jiu Hwang
資訊學院數位圖書資訊學程
關鍵字: 複雜網路;視覺化;圖書借閱;網路基調;Complex Network;Virtualization;Book Circulation;Motif Detection
公開日期: 2008
摘要: 分析圖書館讀者之借閱記錄有助於了解讀者的需求及對圖書館服務滿意的程度,本研究為國內首次以複雜網路理論為基礎,用圖形化網路方法來分析圖書館借閱記錄,透過讀者間借閱的關係,把圖書館讀者之借閱記錄轉化為圖書借閱網路,探討其: (1) 圖書借閱網路基本性質; (2) 圖書借閱網路之趨勢變化; (3) 圖書借閱網路基調(Motif)。 本研究以交通大學圖書館2000年至2007年的圖書借閱記錄為例,由個案研究的結果中發現交大圖書借閱網路具有「小世界」、「無尺度」等複雜網路特性,而從圖書借閱網路趨勢變化分析中也發現了以下現象及特徵: (1) 讀者間借閱關連度很低,存在借閱小團體,小團體間的借閱關連度也很低; (2) 僅有少數圖書被大量的借閱,即館藏使用情形符合80/20定理; (3) 讀者借閱分佈有越來越集中化的趨勢,但同時也有邊緣化現象存在; (4) 隨著時間的變化以上趨勢有越來越明顯的現象。 透過此一網路分析,能夠更進一步的了解讀者借閱行為中的趨勢變化及其意涵,讓圖書館界及相關研究人員能透過視覺化的介面快速得到宏觀的讀者借閱的行為及其趨勢變化,以做為深入研究或圖書館策略制定的參考依據。
Library circulation record analysis helps to understand users’ needs and satisfaction. This research is the first one that uses graphic method to analyze Library circulation records baseing on Complex Network. The researcher built Library circulation network from Library circulation records through the relationship between users and that between users and books. The research attempts the following analyses: (1) Properties of Library circulation network; (2) The phenomenons and trend of Library circulation network; (3) Motif inside Library circulation network. We use Library circulation records between years 2000 and 2007 from NCTU Through the case study, we found that Library circulation network has “Small World” and “Scale-Free network” characteristics. Through analyzing the trend of Library circulation network, we also found the following phenomenons and characteristics: (1) Relationship between users is quite weak, user groups exist. Relationship between groups is quite weak too; (2) Only a few books have large circulation record, which fitswith the 80/20 theory; (3) Distribution of user’s circulation is getting closer by the time, so is Long Tail phenomenon at the same time; (4) The trend is becoming more obvious by the time. Through this network analysis, we can understand more about the trend and meaning from reader’s behavior, Librarians and researchers can gain an overall pciture of user’s behavior and trend through this virtualization interface. The findings can serve as the base for further research or Library strategic making.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009567612
http://hdl.handle.net/11536/39876
顯示於類別:畢業論文