近年來社群網路已經被十分的關注，舉凡一些社群網站Facebook、LinkedIn 以及Google+，人們對於從其中獲得潛在、不是顯而易見的資訊感到興趣，然而這些社群網路問題利用手動分析是非常複雜的。資料探勘是一個廣泛統稱的技術可以解決這些問題。本論文將探討如同社群網路形式的人際關係網，研究學校以及職場兩個人際關係網，利用資料探勘方法裡的聚類分析，觀察影響學校以及職場成員聚落形成的可能原因。

Social network has obtained remarkable attention for the past few years. Accessing social network sites such as Facebook , LinkedIn and Google+. People are interested to get implicit, non-trivial knowledge from the data. However, it is very complex to analyse manually. Data mining provides a wide range of techniques for solving this issue. In this thesis, we study friendship networks based on school and workplace, using cluster analysis of data mining and observing the reason of colony formation.

論文審定書 i
誌謝 ii
摘要 iii
Abstract iv
目錄 v
圖次 vii
表次 ix
第一章 緒論 1
1.1 前言 1
1.2 研究動機 2
1.3 論文架構 3
第二章 文獻探討 4
2.1 社群網路分析 4
2.1.1 社群網路分析概述 4
2.1.2 社群網路分析流程與介紹 4
2.2 聚類分析演算法 6
2.2.1 K-平均演算法 6
2.2.2 鬆弛K-平均演算法 10
2.2.3 譜聚類演算法 12
2.2.4 置信傳播聚類演算法 22
第三章 實驗與結果分析 25
3.1 資料收集與探索 25
3.1.1 資料收集 25
3.1.2 分群數目 28
3.2 實驗分析 30
3.2.1 學校人際關係網 32
3.2.2 職場人際關係網 41
3.2.3 兩實驗之間比較 50
第四章 結論 51
參考文獻 52
附錄一 推廣放寬比率切割最小化問題以分k 群 54
附錄二 網路資料取得與使用 56
附錄三 利用Pajek繪製網路圖 57
附錄四 問卷設計 59

[1] A. Decelle, F. Krzakala, C. Moore, and L. Zdeborová, “Asymptotic analysis of the stochastic block model for modular networks and its algorithmic applications,” Physics Review E, 84:066106, Dec 2011.
[2] M. E. J. Newman, “Finding community structure in networks using the eigenvectors of matrices,” Physical Review E 74, 036104 (2006).
[3] W. W. Zachary, “An information flow model for conflict and fission in small groups,” Journal of Anthropological Research 33, 452–473 (1977).
[4] M. Jordan, J. Kleinberg, and B. Scholkopf, Information Science and Statistics, 1st ed. Cambridge, U.K.: Springer, 2006.
[5] U. Luxburg, “A tutorial on spectral clustering,” Statistics and Computing, 17(4):395, 2007.
[6] A. Singh, “Spectral clustering,” Machine Learning Lectures, CMU, November 2010.
[7] J. S. Jang, “K-means clustering,” Scientific Computing Lectures, NTHU, November 2011.
[8] P. Awasthi, A. S. Bandeira, M. Charikar, R. Krishnaswamy, S. Villar, and R. Ward, “Relax, no need to round: integrality of clustering formulations,” preprint, 2014. [Online]. Available: http://arxiv.org/abs/1408.4045.
[9] T. Hastie, R. Tibshirani, J. Friedman, and J. Franklin, The elements of statistical learning: data mining, Springer-Verlag, 2009.
[10] B. Mohar, “The Laplacian spectrum of graphs,” in Graph Theory, New York, 1991, pp. 871–898.
[11] B. Mohar, “Some applications of Laplace eigenvalues of graphs,” in Graph Symmetry: Algebraic Methods and Applications, Kluwer , 1997, pp. 225–275.
[12] L. Hagen and A. B. Kahng, “New spectral methods for ratio cut partitioning and clustering,” IEEE Transactions on CAD, 11:1074–1085, 1992.
[13] G. Golub and C. van Loan, Matrix Computations. Johns-Hopkins University Press, 1996.