22 Retrieval support system using bookmarks that are shared in an organization 1110250 2011 3 17
IT,, i
Abstract Retrieval support system using bookmarks that are shared in an organization Yoshihiko Komaki In an organization, it is very important that we make use of the information using personal knowledge in the organization in order to obtain knowledge. Since early times, various forms of information sharing is done. In recently, the method which efficiently shares information is proposed by information sharing system using IT tecnology. In existing research, there is a system recommended shared bookmark to the hobbies and diversions of the user. As a problems, shared bookmark is not classified. So right recommendation is not performed. In this thesis, developed a system classifies shared bookmark automatically, and developed and suggested a system more recommended shared bookmark to the hobbies and diversions of the user. We compared the suggestion system with the existing system. As a result, suggested system was better than an existing system. key words Information sharing, retrieval support, Social Bookmark ii
1 1 1.1.................................. 1 1.2................................. 2 2 3 2.1.............................. 3 2.2............................. 3 2.2.1............................... 4 2.2.2................................. 4 2.2.3............................. 4 2.3....................... 5 3 7 3.1 Web............. 7 3.2................................. 8 3.2.1...................... 8 3.2.2........................... 9 3.2.3.......................... 9 3.2.4.......................... 9 3.3........................... 10 3.3.1.................................. 10 3.3.2................................. 10 4 12 4.1................................. 12 iii
4.2................................. 13 4.3............... 13 4.3.1...................... 14 4.3.2................................ 15 4.4............................. 17 5 19 5.1.................................. 19 5.2............................. 19 5.3...................................... 20 6 22 23 24 iv
2.1............................ 4 2.2......................... 6 3.1............................... 7 3.2............................ 10 4.1............................... 13 4.2...................... 16 4.3............................ 18 5.1...................... 21 v
3.1............................... 8 3.2................................. 9 4.1 p(d i z k ).................... 15 4.2 PHP.............. 16 5.1................................... 20 vi
1 1.1 Web [1] Web Google Web 1998 2008 2600 1 [2] Web RSS Web Web RSS Web RSS [3] 1 [4] 1
1.2 1.2 2
2 2.1 IT IT 2.1 2.2 IT 3 3
2.2 2.1 2.2.1 2.2.2 2.2.3 Web 4
2.3 2.3 1 Web Livedoor 2.2 5
2.3 2.2 6
3 [5] 3.1 Web Web 3.1 3.1 4 7
3.2 3.2 3.2.1 3.1 ID URL 3.1 ID URL A http://www.xxxx... PHP Java... A http://www.xxyy... Java eclipse SQL... B http://www.xyzx...... C http://www.zzzz... Apple Mac ipod itunes............ 8
3.2 3.2.2 ID 30 3.2 ID 3.2 ID A PHP JavaScript SQL MySQL Apache Java... B... C Java eclipse......... 3.2.3 30 14 14 3.2.4 3.2 9
3.3 3.2 3.3 3.3.1 7 30 (3.1) = 30 100 (3.1) 7 47.6% 3.3.2 10
3.3 PHP PHP 11
4 4.1 4.1 3 Livedoor [6] 12
4.2 4.1 4.2 3.2.1 3.2.3 4.3 13
4.3 [7] 4.3.1 Probabilistic Latent Semantic Indexing PLSI [8] PLSI 1 d w z 4 p(d, w) = z p(z)p(w z)p(d z) (4.1) D = (d 1,d 2,...,d N ) D W = (w 1,w 2,...,w M ) (4.2) L = N M n(d i, w j )logp(d i, w j ) (4.2) i=1 j=1 EM E-step Q (t+1) ijk = p(z k ) (t) p(w j z k ) (t) p(d j z k ) (t) K k=1 p(z k) (t) p(w j z k ) (t) p(d j z k ) (t) (4.3) M-step p(w j z k ) (t) = N i=1 n(d i, w j )Q (t) ijk N i=1 M j=1 n(d i, w j )Q (t) ijk (4.4) 14
4.3 p(z k ) (t) = K k=1 N i=1 M j=1 n(d i, w j )Q (t) ijk N i=1 M j=1 n(d i, w j )Q (t) ijk (4.5) p(d i z k ) (t) = M j=1 n(d i, w j )Q (t) ijk N i=1 M j=1 n(d i, w j )Q (t) ijk (4.6) E-step M-step p(w j z k ) d i z k p(d i z k ) p(d i z k ) 4.1 4.1 p(d i z k ) 文書 クラスタ クラスタ 1 クラスタ 2 クラスタ 3 クラスタ 4 文書 1 0.423 0.121 0.057 0.038 文書 2 0.341 0.098 0.008 0.002 文書 3 0.102 0.012 0.311 0.008 文書 4 0.001 0.021 0.021 0.417 文書 5 0.007 0.333 0.102 0.135 文書 6 0.051 0.101 0.271 0.019 1 1 2 2 5 3 3 6 4 4 4.3.2 PHP PHP 15
4.3 PHP PHP (A,B,C,D,...) 4.2 4.2 PHP 1 A G L O... 2 D H J N... 3 B E F M... 4 C I K P... 4.2 4.2 16
4.4 4.4 4 PHP PHP PHP PHP 4 4 PHP PHP 4.3 17
4.4 4.3 18
5 10 5.1 3 PHP 30 PHP 38 PHP 30 98 (5.1) = 100 (5.1) PLSI 20 5.1 87.5% 5.2 10 30 5.1 19
5.3 5.1 PHPに関するページ 1 回目 2 回目 4 回目 5 回目 6 回目 7 回目 8 回目 9 回目 10 回目 動作サンプル 71.8 92.3 59 87.2 94.9 64.1 94.9 92.3 94.9 プログラム 92.6 92.6 92.6 92.6 92.6 92.6 92.6 92.6 92.6 概要 説明 62.5 87.5 84.4 87.5 87.5 84.4 87.5 87.5 87.5 PHPに関するページ 11 回目 12 回目 14 回目 15 回目 16 回目 17 回目 18 回目 19 回目 20 回目 動作サンプル 92.3 51.3 89.7 92.3 89.7 89.7 94.9 87.2 89.7 プログラム 92.6 92.6 92.6 92.6 92.6 92.6 92.6 92.6 92.6 概要 説明 87.5 71.9 90.6 87.5 87.5 87.5 87.5 87.5 87.5 (5.2) = 30 100 (5.2) 10 71.3% 5.3 87.5% 47.6% 71.3% PLSI 20
5.3 5.1 21
6 PLSI PLSI 47.6% 72.6% 1 1 1 22
23
[1] - Enterprise 2.0 - ZDNet Japan, http://japan.zdnet.com/sp/feature/07sp0060/story/0,3800076669,20346851,00.htm, 2007. [2] Jesse Alpert and Nissan Hajaj. We knew the web was big.... Official Google Blog. 2008-07-25. http://googleblog.blogspot.com/2008/07/we-knew-webwas-big.html, 2007. [3] Karger,D.R. and Quan,D., What would it mean to blog on the semantic web?, in Proc. Third International Semantic Web Conference, pp.214-228,2004. [4],, 19, 2008 [5], Web,,, 2010. [6] Livedoor 2008 12, http://labs.edge.jp/datasets/ 2009. [7] Ver.2.3.3,,http://chasen.naist.jp/hiki/ChaSen/, 2003. [8] Thomas Hofmann, Probabilistic Latent Semantic Indexing, the 22nd annual international ACM SIGIR conference on Research and development in infomation retrieval, ACM, pp.50-57, 1999. 24