近年來隨著網際網絡爆發性的發展，在網路網絡中的訊息量也快速成長，人
們的交際關係從現實環境中逐漸轉變成為虛擬的網際網絡。根據統計，知名交友
網站MySpace在2006年9月就有一億六千萬註冊的成員在上面彼此交流。因此在做
推薦中如何滿足於個人需求在推薦系統的發展中獲得愈來愈多人的關注與重視。
我們發現一個使用者在網絡中的決定除了本身主動決定他與其他使用者的網絡關
係外，該使用者也受到與該使用者間連結的網際網絡結構所影響。如何將網絡的
結構屬性帶入推薦系統中，並且促使推薦能夠滿足個人於個人需求外，並會影響
整體網絡變化，使得整體網絡結構在推薦過程中能夠更加緊密。因此我們將社會
網絡(Social Network)的結構屬性加入推薦系統中。本論文先將使用者在網絡中的
訊息量做為人與人關係的比重，接著我們將使用者在網絡架構中的屬性加入考
量。在網絡假如積極發表回應的使用者我們給予較高的機會推薦給其他使用者，
也考量了使用者與使用者之間可能擁有的共同關係。將網絡架構的屬性、積極的
使用者與共同關係做整體的規劃，並且推薦出較適合使用者與網絡發展的推薦給
予行動者。在實驗中，我們可以知道網絡的平均距離與推薦給使用者的準確度都
優於單純使用訊息量為比重的推薦系統(WMR)。

In recent years, the information on the web grows explosively. More and more people
make friends on the website. According to the statistics, the My Space has 160 mil-
lion registered users until September 2006. The effective recommendation in a large
network attracts more and more attention in the development of recommendation
system. We observed that users in the network make friends based on their inter-
ests or favor. Their actions are also affected by the network structure. We further
consider the features of network structure into the recommendation system. In this
thesis, we regard the message ratio between users as the strength of the relation.
Further, we add the social network features into the recommendation system. If the
user have higher comment ratio, they have higher probability to be recommended
to users. Besides, we further consider common relations between the user and other
members. We combine these factors into the recommendation system. The personal
requirement of recommendation is satisfied in our recommendation system. The
network structure is also tighter. Our experiments show that the average distance
of the network and the accuracy of recommendation are better than WMR which is
a recommendation system based on message ratio.

[1] G. Adomavicius and A. Tuzhilin, “Toward the next generation of recommender
systems: A survey of the state-of-the-art and possible extensions,” in the pro-
ceedings of Knowledge and Data Engineering, IEEE Transactions on, vol. 17,
no. 6, pp. 734–749, 2005.
[2] R. J. Mooney and L. Roy, “Content-based book recommending using learning
for text categorization,” in Proceedings of DL-00, 5th ACM Conference on Dig-
ital Libraries, (San Antonio, US), pp. 195–204, ACM Press, New York, US,
2000.
[3] M. J. Pazzani, “A framework for collaborative, content-based and demographic
filtering,” Artif. Intell. Rev., vol. 13, no. 5-6, pp. 393–408, 1999.
[4] S. Milgram, “The small world problem,” Psychology Today, vol. 2, pp. 60–67,
1967.
[5] G. Palla, A.-L. Barabasi, and T. Vicsek, “Quantifying social group evolution,”
Nature, vol. 446, pp. 664–667, April 2007.
[6] C. Lam, “Snack: incorporating social network information in automated col-
laborative filtering,” in EC ’04: Proceedings of the 5th ACM conference on
Electronic commerce, (New York, NY, USA), pp. 254–255, ACM, 2004.
[7] D. W. Mcdonald, “Recommending collaboration with social networks: a com-
parative evaluation,” in CHI ’03: Proceedings of the SIGCHI conference on
Human factors in computing systems, (New York, NY, USA), pp. 593–600,
ACM Press, 2003.
[8] S. Lo and C. Lin, “Wmr–a graph-based algorithm for friend recommendation,”
in the proceedings of Web Intelligence, 2006. WI 2006. IEEE/WIC/ACM In-
ternational Conference on, pp. 121–128, Dec. 2006.
[9] A. Rapoport, “Contribution to the theory of random and biased nets,” Bulletin
of Mathematical Biology, vol. 19, pp. 257–277, Dec. 1957.
44[10] R. A. Hanneman, “Introduction to Social Network Methods,” University of
California, 2001.
[11] M. E. J. Newman, “The structure and function of complex networks,” In Pro-
ceedings of the Society for Industrial and Applied Mathematics(SIAM), Review
45, vol. 45, March 2003.
[12] J. P. Scott, Social Network Analysis: A Handbook. SAGE Publications, January
2000.
[13] S. Wasserman, K. Faust, and D. Iacobucci, Social Network Analysis : Meth-
ods and Applications (Structural Analysis in the Social Sciences). Cambridge
University Press, November 1994.
[14] T. Lonier and C. H. Matthews, “Measuring the impact of social networks on
entrepreneurial success: The master mind principle,” Presented at the 2004
Babson Kauffman Entrepreneurship Research Conference., 2004.
[15] L. Backstrom, D. Huttenlocher, J. Kleinberg, and Et, “Group formation in
large social networks: Membership, growth, and evolution,” In Proceedings of
the ACM SIGKDD International Conference on Knowledge Discovery and Data
Mining, pp. 44–54.
[16] J. S. Breese, D. Heckerman, and C. Kadie, “Empirical analysis of predictive
algorithms for collaborative filtering,” in Proceedings of the 14th Conference on
Uncertainty in Artificial Intelligence, pp. 43–52, 1998.
[17] H. C. Garton, L. and B. Wellman, “Studying online social networks,” in the
proceedings of Journal of Computer-Mediated Communication. SAIS Review,
vol. 22, pp. 23–37, 1997.
[18] D. Liben-Nowell and J. Kleinberg, “The link-prediction problem for social net-
works,” in the proceedings of Journal of the American Society for Information
Science and Technology, vol. 58, no. 7, pp. 1019–1031, 2007.
[19] G. Salton and M. J. Mcgill, Introduction to Modern Information Retrieval. New
York, NY, USA: McGraw-Hill, Inc., 1986.
45
[20] G. Jeh and J. Widom, “Simrank: a measure of structural-context similarity,”
in KDD ’02: Proceedings of the eighth ACM SIGKDD international conference
on Knowledge discovery and data mining, (New York, NY, USA), pp. 538–543,
ACM Press, 2002.
[21] M. A. Hasan, V. Chaoji, S. Salem, and M. Zaki, “Link prediction using super-
vised learning,” in In Proc. of SDM 06 workshop on Link Analysis, Countert-
errorism and Security, 2006.
[22] L. Ngo and P. Haddawy, “Probabilistic logic programming and bayesian net-
works,” in In Asian Computing Science Conference, vol. 1023, pp. 286–300,
Springer-Verlag, 1995.
[23] N. Friedman, L. Getoor, D. Koller, and A. Pfeffer, “Learning probabilistic
relational models,” in In Proceedings of the Sixteenth International Joint Con-
ference on Artifical Intelligence, pp. 1300–1309, Springer-Verlag, 1999.
[24] L. Getoor, N. Friedman, D. Koller, and B. Taskar, “Learning probabilistic
models of link structure,” J. Mach. Learn. Res., vol. 3, pp. 679–707, 2003.
[25] R. van Meteren. and M. van Someren, “Using content-based filtering for rec-
ommendation,” ECML/MLNET Workshop on Machine Learning and the New
Information Age, Barcelona, pp. 47–56, 2000.
[26] L. Adamic and E. Adar, “Friends and neighbors on the web,” Social Networks,
vol. 25, no. 3, pp. 211–230, 2003.
[27] A. Kaban and X. Wang, “Context based identification of user communities from
internet chat,” Neural Networks, 2004. Proceedings. 2004 IEEE International
Joint Conference on, vol. 4, pp. 3287–3292 vol.4, July 2004.
[28] B. Chen, Q. Zhao, B. Sun, and P. Mitra, “Predicting blogging behavior using
temporal and social networks,” in the proceedings of Data Mining, 2007. ICDM
2007. Seventh IEEE International Conference on, pp. 439–444, Oct. 2007.
[29] K. Iwahama, Y. Hijikata, and S. Nishida, “Content-based filtering system for
music data,” Applications and the Internet Workshops, 2004. SAINT 2004
Workshops. 2004 International Symposium on, pp. 480–487, Jan. 2004.
46
[30] J. Palau, M. Montaner, B. L′opez, and J. L. de la Rosa, “Collaboration anal-
ysis in recommender systems using social networks,” in the proceedings of the
Cooperative Information Agents(CIA), pp. 137–151, 2004.
[31] K.-Y. Jung, D.-H. Park, and J.-H. Lee, “Personalized movie recommender sys-
tem through hybrid 2-way filtering with extracted information,” in Flexible
Query Answering Systems(FQAS), pp. 473–486, 2004.
[32] P. Kazienko and K. Musial, “Mining personal social features in the community
of email users,” in In the proceedings of the 2008 SOFSEM theory and practice
of computer science, pp. 708–719, 2008.
[33] J. Karamon, Y. Matsuo, H. Yamamoto, and M. Ishizuka, “Generating social
network features for link-based classification,” in PKDD 2007: Proceedings of
the 11th European conference on Principles and Practice of Knowledge Discov-
ery in Databases, (Berlin, Heidelberg), pp. 127–139, Springer-Verlag, 2007.
[34] S. Q. Q. L. Qihong Liu, Changjie Tang and F. Wen, “Mining the core member
of terrorist crime group based on social network analysis,” in the proceedings of
Intelligence and Security Informatics, vol. 4430, pp. 311–313, 2007.
[35] W. Buntine and T. Niblett, “A further comparison of splitting rules for decision-
tree induction,” Machine Learning, vol. 8, no. 1, pp. 75–85, 1992.
[36] R. Nolker and L. Zhou, “Social computing and weighting to identify member
roles in online communities,” in the proceedings of Web Intelligence, 2005. Pro-
ceedings. The 2005 IEEE/WIC/ACM International Conference on, pp. 87–93,
Sept. 2005.
[37] S. Borgatti, “Netdraw,” www.analytictech.com, 2008.
[38] J. L. Herlocker, J. A. Konstan, L. G. Terveen, and J. T. Riedl, “Evaluating
collaborative filtering recommender systems,” In the proceedings of the ACM
transactions on information systems (TOIS), vol. 22, pp. 5–53, January 2004.
[39] B. Sarwar, G. Karypis, J. Konstan, and J. Riedl, “Analysis of recommendation
algorithms for e-commerce,” in EC ’00: Proceedings of the 2nd ACM conference
on Electronic commerce, (New York, NY, USA), pp. 158–167, ACM Press, 2000.