近年來，圖書自動化管理系統已成為圖書館館際合作相關作業極為重要的一環，而跨區域、跨縣市聯合書目檢索的功能更是圖書自動化管理系統中極其重要的部分。然而有些書目可能因為不同圖書館的編目習慣不一致、或對於編目標準的認定不同、又或輸入錯誤等，直接以詞彙為書目特徵，無法有效檢索出不完全相同詞彙的近似書目。本研究以「文獻機讀編目格式」為依據，將擁有全國圖書館書目的國家圖書館之原始書目資料透過Bigram索引與權重計算抽取出具作為分辨書目代表性的關鍵書目特徵。

本研究為了發掘各書目間之相似關係程度，先利用向量空間模型相似度之計算與二分群集法進行相似書目之分群；然後再透過粒子群最佳化群集法，以期能求得全域最佳解，獲得更好的書目分群品質。書目群集除了可加速近似書目檢索，亦可進一步將每一本書分散於各個圖書館，編目不盡相同的所有書目比對出來，並統整記錄在一起，將能大大提升書目檢索的效率。

In recent years, the library automation system has become an important part of library inter-cooperation, and cross-regional, cross-county municipal joint bibliographic search function is an extremely important part. However, some bibliographic may be due to different habits of library cataloging inconsistent, or the different standards, or typing errors, etc., user can't directly use the bibliographic keyword to find the similar but not identical bibliographic. In this study, We the basis CMARC format based on the original bibliographic data of the National Library of Bigram indexes and the weighted calculation extracted has to distinguish between a bibliographic representative bibliography characteristics. We use the Vector Space Model to find the similarity relations between all bibliographic through Bisecting K-means Clustering approximation of the clustering of the bibliographic, and then hope to obtain better clustering quality through the PSO clustering. The bibliography cluster can accelerate approximate bibliographic retrieval, also can find scattered in various library cataloging is not exactly the same bibliographic and then compiled these record together, so it can effectively improve the efficiency of bibliographic retrieval.

摘要i
Abstractii
誌謝iii
目錄iv
圖目錄vi
表目錄vii
第一章 緒論1
1.1 研究背景1
1.2 研究動機與目的1
1.3 論文架構2
第二章 文獻探討3
2.1文獻分析機讀格式3
2.2文件特徵5
2.2.1關鍵詞頻率(Term Frequency, tf)5
2.2.2反向文件頻率(Inverse Document Frequency, idf)6
2.2.3 Bigram索引(Bigram indexes)6
2.3資訊檢索向量空間模型(Vector Space Model in Information Retrieval) 8
2.4 二分群集法(Bisecting K-means Clustering)8
2.5 粒子群最佳化(Particle Swarm Optimization, PSO)9
2.6 檢索效能評估11
第三章 研究方法13
3.1 研究流程13
3.2 書目前置處理14
3.2.1 原始書目欄位拆解15
3.2.2 過濾書目欄位字元18
3.3 特徵抽取19
3.4 相似書目分群一: 二分群集法20
3.5 相似書目分群二: 粒子群最佳化群集法22
3.6 書目檢索23
第四章 實驗結果與評估26
4.1 書目資料26
4.2 實驗設計與結果26
4.2.1實驗一：以斷詞為特徵的書目檢索26
4.2.2實驗二：以Bigram索引為特徵的書目檢索30
4.2.3實驗三：以Bisecting K-means+PSO分群後，提升檢索效率33
4.3 錯誤分析39
4.3.1 錯誤類型39
4.3.2 錯誤改善39
第五章 結論與建議 41
參考文獻 42

[1]文獻分析機讀格式修訂小組，文獻分析機讀格式，國家圖書館，民國90年。
[2]Christoper D. Manning, Prabhakar Raghavan, and Hinrich Schutze, Introduction to Information Retrieval, Cambridge University Press, 2008.
[3]G. Salton, A. Wong, and C. S. Yang, "A Vector Space Model for Automatic Indexing," Communications of the ACM, Vol. 18, No.11, 1975, pp.613-620.
[4]DW van der Merwe and AP Engelhrecht, "Data Clustering using Particle Swarm Optimization," IEEE Congress on Evolutionary Computation, 2003, pp.215-220.
[5]James Kennedy and Russell Eberhart, "Particle Swarm Optimization," IEEE International Conference on Neural Network, Vol.4, 1995, pp. 1942-1948.
[6]Yuhui Shi and Russell C. Eberhart, "Empirical Study of Particle Swarm Optimization," IEEE Congress on Evolutionary Computation, 1999, pp.1945-1950.
[7]Xiaohui Cui, and Thomas E. Potok, "Document Clustering Analysis Based on Hybrid PSO+K-means Algorithm," Journal of Computer Science, Special issue on Efficient Heuristics for Information Organization, Science Publications, New York, 2005, pp. 27-33.
[8]余顯強, "應用延伸標示語言與圖書館機讀格式轉換之研究," 2004年現代資訊組織與檢索研討會, Nov. 2004.
[9]張慧銖、郭蕙貞, "大學圖書館書目紀錄品質研究-以台灣六所大學為例", 中興大學圖書資訊研究所碩士論文, 民國99年.
[10]Jones, K.S., “A Statistical Interpretation of Terms Specificity and its Application in Retrieval,” Journal of Documentation, Vol. 28, No. 5, pp.111-121, 1972.
[11]Alan G. Konheim, Cryptography: A Primer, JohnWiley & Sons, 1981.
[12]Ken Lunde, CJKV Information Processing, O'Reilly, 1998.
[13]Hollink, Vera, Jaap Kamps, Christof Monz, and Maarten de Rijke. 2004, Monolingual document retrieval for European languages.
[14]Tomlinson, Stephen. 2003. Lexical and algorithmic stemming compared for 9 European languages with Hummingbird Searchserver at CLEF 2003. In Proc. Cross-Language Evaluation Forum, pp.286-300.
[15]鄭貞信、徐瑞杰、李宏林、謝淑諭、葉冠群、邊國維, "英中日韓文的跨語言檢索之比較,"DTIM 2007 中華民國數位科技與創新管理研討會,June 2007.
[16]Michael Steinbach, George Karypis, and Vipin Kumar, "A comparison of Document Clustering Techniques," KDD-2000, Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, August 20-23, 2000.
[17]楊燕珠、陳寶燦, "應用分群技術於同義書目之過濾與最佳化,"CSMOT 2009 中華民國科技管理學會年會暨論文研討會, Dec. 2009.