untitled

Size: px

Start display at page:

Download "untitled"

依马佳
4 years ago
Views:

1 SP-I 2010 SNS Comparison Analysis among Large Amount of SNS Sites Fujio Toriumi Hitoshi Yamamoto Hirohiko Suwa Isamu Okada Kiyoshi Izumi Yasuhiro Hashimoto Graduate School of Information Science, Nagoya University tori@is.nagoya-u.ac.jp, tori/ Facalty of Business Administration, Rissho University hitoshi@ris.ac.jp, Graduate School of Information Systems, University of Electro-Communications h-suwa@is.uec.ac.jp, Faculty of Business Administration, Soka University okada@soka.ac.jp, Digital Human Research Center, AIST kiyoshi@ni.mints.ne.jp, Department of Systems Innovation, School of Engineering, The University of Tokyo hy@sys.t.u-tokyo.ac.jp, keywords: social network services, network analysis, sns Summary In recent years, application of Social Networking Services (SNS) and Blogs are growing as new communication tools on the Internet. Several large-scale SNS sites are prospering; meanwhile, many sites with relatively small scale are offering services. Such small-scale SNSs realize small-group isolated type of communication while neither mixi nor MySpace can do that. However, the studies on SNS are almost about particular large-scale SNSs and cannot analyze whether their results apply for general features or for special characteristics on the SNSs. From the point of view of comparison analysis on SNS, comparison with just several types of those cannot reach a statistically significant level. We analyze many SNS sites with the aim of classifying them by using some approaches. Our paper classifies 50,000 sites for small-scale SNSs and gives their features from the points of network structure, patterns of communication, and growth rate of SNS. The result of analysis for network structure shows that many SNS sites have small-world attribute with short path lengths and high coefficients of their cluster. Distribution of degrees of the SNS sites is close to power law. This result indicates the small-scale SNS sites raise the percentage of users with many friends than mixi. According to the analysis of their coefficients of assortativity, those SNS sites have negative values of assortativity, and that means users with high degree tend to connect users with small degree. Next, we analyze the patterns of user communication. A friend network of SNS is explicit while users communication behaviors are defined as an implicit network. What kind of relationships do these networks have? To address this question, we obtain some characteristics of users communication structure and activation patterns of users on the SNS sites. By using new indexes, friend aggregation rate and friend coverage rate, we show that SNS sites with high value of friend coverage rate activate diary postings and their comments. Besides, they become activated when hub users with high degree do not behave actively on the sites with high value of friend aggregation rate and high value of friend coverage rate. On the other hand, activation emerges when hub users behave actively on the sites with low value of friend aggregation rate and high value of friend coverage rate. Finally, we observe SNS sites which are increasing the number of users considerably, from the viewpoint of network structure, and extract characteristics of high growth SNS sites. As a result of discrimination on the basis of the decision tree analysis, we can recognize the high growth SNS sites with a high degree of accuracy. Besides, this approach suggests mixi and the other small-scale SNS sites have different character trait. 1. Social Networking Service: SNS Blog SNS SNS

2 SNS 79 SNS SNS SNS SNS SNS SNS SNS SNS So-net SNS () 1 So-net SNS () SNS SNS SNS SNS ( So-net) 50,000 SNS SNS SNS SNS SNS 20% SNS [int 08] MySpace 2 2 mixi 3 1,500 Gree SNS SNS [Adamic 03] Cyworld 5 1,000 [Ahn 07] mixi [Yuta 07] mixi [ 06] [ 07] SNS SNS SNS SNS SNS SNS SNS SNS NPO SNS[ 07] SNS[ 07] SNS SNS SNS mixi MySpace SNS OpenPNE 6 SNS SNS SNS SNS SNS 2 SNS 3 SNS SNS 4 SNS SNS SNS SNS SNS SNS SNS SNS SNS SNS SNS SNS SNS

3 SP-I SNS 6636 SNS SNS % 100 SNS % 1 SNS SNS SNS SNS 6636 SNS SNS mixi [Yuta 07] Cyworld [Ahn 07] SNS [Ishida 08] SNS SNS SNS SNS SNS SNS So-net SNS () SNS 1 [Watts 98] SNS L L SNS SNS % SNS 1.9 L 2.1 SNS C SNS SNS 0.1 C % SNS 0.2 SNS L C L C L C SNS L C L SNS C % SNS

4 SNS 81 So-net SNS () SNS SNS 2 log(y) R SNS SNS R 2 > 0.6 SNS 409 SNS γ R 2 =0.877,γ= SNS 4 mixi -2.4 [ 07] mixi mixi % SNS 3 4 (R 2 =0.877,γ = 1.361) 2 [Barabási 99] SNS R 2 i R 2 =1 (log(y i) log(f i )) 2 i (log(y (1) i) log(y)) 2 log(y i ) log(f i ) 3 (Assortativity)r[Newman 02] r = M 1 i j ik i [M 1 i 1 2 (j i + k i )] 2 M 1 i (j2 i + k2 i ) [M 1 i 1 2 (j i + k i )] 2 (2) 1 r 1 r r SNS

5 SP-I 2010 mixi r =0.121[Yuta 07] Cyworld r = 0.13[Ahn 07] r SNS SNS 3 0.5% % SNS So-net SNS () SNS 1 SNS SNS 2.0 C 0.0 C 0.8 SNS 1 L C R 2 γ ( ) r ( ) γ R 2 > 0.6 SNS K-Means C1 C5 SNS 2 C1 SNS SNS SNS C2 C1 SNS C1 SNS C3 C SNS R 2 C3 SNS C3 SNS C4 C SNS L SNS C4 SNS C4 SNS SNS SNS C5 C1 L So-net SNS () L 2 C5 SNS 2 SNS 2 4 SNS SNS SNS K-Means k =5 L C R 2 r SNS C C C C C

6 SNS SNS SNS SNS SNS , SNS 100 SNS 5 SNS 3 1 = = 3 2 SNS ( 5) =0.737 =0.610 (3) (4) 3 3 SNS SNS 3 N SNS SNS SNS

7 SP-I N F (x) (x) 1.795(f).946(f) (f) 4.846(f) 5.525(f).001***.000*** ***.003**.001** p<.001, p<.01 (x) 2, (f) F 4 PC N ***.000***.020*.010* p<.001, p< SNS (PC) 4 SNS SNS SNS 3 5 SNS SNS SNS SNS SNS SNS 5 SNS SNS SNS SNS

8 SNS 85 5 PC N= N= N=81.018* * N= p<.05, +p< SNS 4 1 SNS SNS SNS SNS SNS SNS 3 ( ) % SNS 3 20% 20% ( ) 3 20% SNS () SNS SNS 100 SNS SNS (3 20% ) SNS 19.86% SNS SNS 100 SNS 3 20 SNS 7 SNS 8 7 () SNS C4.5 J SNS % 45.9% F (sum), SNS (Day), (active7, active3, active1), (type), (avgf riend) (createt ype=pc,mobile), (PV), PC (pv pc, pv ktai), PC (pv admin, pv mobileadmin), PC (uu pc, uu ktai). 8 (LinkNum), (AverageP ath), (ClusterV alue), (DegreeP owerindex), (MeanDegree), (MaxDegree), (Assortativity), (DegreeZeroN um) 2 9 C4.5 WEKA[Weka, Witten 05]

9 SP-I % 97.0% F % 45.1% F % 99.8% F SNS % 42.5% F % 94.8% F SNS SNS 4 SNS (Day) (1) 1 (Day 39) SNS SNS 3 (MeanDegree) 2.64 SNS SNS 2.64 SNS 25% ( 7(a)) (2) 1 (39 <Day 143) SNS 25% SNS SNS (uu pc 70, DegreeZeroNum 0.01, avgf riend > 2.04) 88.23% ( 7(b1)) SNS(uu pc > 70, ClusterV alue > 0.15) ( 7(b2)) (3) 1 (143 <Day 323) SNS SNS (sum > 922 type = open MeanDegree 6.60) 75% ( 7(c)) (4) 1 (323 <Day) SNS 4.6% ( 7(d)) 5. SNS SNS 5 1 SNS SNS SNS 1 6 mixi [ 07] [Yuta 07] 1 So-net SNS () R 2 6 L γ r mixi So-net SNS () mixi SNS 6 mixi L C γ r mixi So-net SNS () mixi [ 07, Yuta 07] 2 mixi mixi IT [ 06] mixi 3 10

10 SNS 87 7 : T 3F mixi mixi SNS 3 SNS mixi mixi % 1 SNS 20% mixi So-net SNS () SNS mixi SNS So-net SNS () SNS 4 So-net SNS () SNS SNS mixi mixi SNS mixi SNS SNS mixi SNS mixi SNS 8 mixi ( ) 5 2 SNS 2 4 SNS SNS So-net SNS () SNS(snsA, snsb) SNS (2 2) (3 5) (4 7) 7 SNS SNS 7 1

01 9 SNS A SNS B 7 SNS snsa snsb C2 C1 d c 11 [Newman 04] 9 2008 4 1 2008 10 1

11 SP-I 2010 SNS A 500 SNS B SNS A SNS B 7 SNS snsa snsb C2 C1 d c 11 [Newman 04] [Palla 07] snsa 3,457 snsb 3,523 SNS 1 snsa 3.1 / 6.5 snsb 4.5 / 4.2 snsa snsb 11 () SNS SNS SNS SNS 6. 50,000 SNS SNS SNS SNS SNS SNS mixi SNS SNS SNS SNS SNS SNS SNS SNS SNS

SNS 89 So-net SNS () [Adamic 03] Adamic, L., Buyukkokten, O., and Adar, E.: A social network caught in the Web, First Monday, Vol. 8, No. 6, p. 29 (2003) [Ahn 07] Ahn, Y., Han, S.

: Analysis of topological characteristics of huge online social networking services, Proceedings of the 16th international conference on World Wide Web, pp.

509 (1999) [int 08] 2008, (2008) [Ishida 08] Ishida, K., Toriumi, F., and Ishii, K.: Proposal for Growth Model of Social Network Services (2008) [Newman 02] Newman, M.

: Fast algorithm for detecting community structure in networks, Physical Review E, Vol. 69, No. 6, p. 066133 (2004) [Palla 07] Palla, G., Barabási, A.-L., and Vicsek, T.

6684, pp. 409 10 (1998) [Weka] Weka http://www.cs.waikato.ac.nz/ ml/weka/ [Witten 05] Witten, I. H. and Frank, E.

12 SNS 89 So-net SNS () [Adamic 03] Adamic, L., Buyukkokten, O., and Adar, E.: A social network caught in the Web, First Monday, Vol. 8, No. 6, p. 29 (2003) [Ahn 07] Ahn, Y., Han, S., Kwak, H., Moon, S., and Jeong, H.: Analysis of topological characteristics of huge online social networking services, Proceedings of the 16th international conference on World Wide Web, pp (2007) [Barabási 99] Barabási, A. and Albert, R.: Emergence of Scaling in Random Networks, Science, Vol. 286, No. 5439, p. 509 (1999) [int 08] 2008, (2008) [Ishida 08] Ishida, K., Toriumi, F., and Ishii, K.: Proposal for Growth Model of Social Network Services (2008) [Newman 02] Newman, M.: Assortative Mixing in Networks, Physical Review Letters, Vol. 89, No. 20, p (2002) [Newman 04] Newman, M.: Fast algorithm for detecting community structure in networks, Physical Review E, Vol. 69, No. 6, p (2004) [Palla 07] Palla, G., Barabási, A.-L., and Vicsek, T.: Quantifying social group evolution, Nature, Vol. 446, pp (2007) [Watts 98] Watts, D. and Strogatz, S.: Collective dynamics of smallworld networks., Nature, Vol. 393, No. 6684, pp (1998) [Weka] Weka ml/weka/ [Witten 05] Witten, I. H. and Frank, E.: Data Mining: Practical Machine Learning Tools and Techniques, Morgan Kaufmann Series in Data Management Systems, Morgan Kaufmann, second edition (2005) [Yuta 07] Yuta, K., Ono, N., and Fujiwara, Y.: A Gap in the Community-Size Distribution of a Large-Scale Social Networking Site, Arxiv preprint physics/ (2007) [ 07], ACS:,, Vol. 48, No. 7, pp (2007) [ 07],,, SNS Web2.0, (2007) [ 07], SNS mixi,, Vol. 22, No. 5 (2007) [ 06] IT, mixi, (2006) [ 06], SNS,,vol47,No.9 (2006) ( ()) () ( ) ()

untitled

untitled LBS Research and Application of Location Information Management Technology in LBS TP319 10290 UDC LBS Research and Application of Location Information Management Technology in LBS , LBS PDA LBS