DEIM Forum 2012 E8-4 Wikipedia y

Size: px

Start display at page:

Download "DEIM Forum 2012 E8-4 Wikipedia y"

せせらかたいわ
5 years ago
Views:

1 DEIM Forum 2012 E8-4 Wikipedia Wikipedia Wikipedia Wikipedia Extracting Difference Information from Multilingual Wikipedia in Consideration of the Quality of Information Yuya FUJIWARA, Yu SUZUKI, Yukio KONISHI, and Akiyo NADAMOTO Graduate School of natural science graduate course, Konan University Graduate School8-9-1 Okamoto Higashinada, Kobe, Hyogo, Japan Information Technology Center, Nagoya University Furotyo, Chikusa, Nagoya, Aichi, Japan Konan International Exchange Center,Konan University8-9-1 Okamoto Higashinada, Kobe, Hyogo, Japan Dept.of Intelligene and Informatics, Konan University8-9-1 Okamoto Higashinada, Kobe, Hyogo, Japan Key words Wikipedia Multilingual Difference Information Quality of Information 1. Wikipedia Wikipedia

2 Wikipedia Wikipedia Bowls Wikipedia Wikipedia Wikipedia Wikipedia 2 Wikipedia Wikipedia Wikipedia 2 n1.htm Wikipedia Wikipedia (2) 4 (3) (2) 5 (4) 6 (5) 2. Wikipedia Michael [1] Milne [2] Wikipedia [3] Wikipedia pfibf Wikipedia pfibf [4] Wikipedia Brin [5] PageRank Jaap [6] Web Wikipedia Web Wikipedia Wikipedia David [7] Wikipedia

Fig. 1 1 Analysis of the link structure Chen [8] Web Web Web Wikipedia Wikipedia Eytan [9] Wikipedia infobox infobox Wikipedia Wikipedia Wikipedia Wöhner [10] Hu [11] 3.

3 Fig. 1 1 Analysis of the link structure Chen [8] Web Web Web Wikipedia Wikipedia Eytan [9] Wikipedia infobox infobox Wikipedia Wikipedia Wikipedia Wöhner [10] Hu [11] Wikipedia Bowling green 1 2 Fig. 2 relevance degreee 2 3 (2) ( 1) 4 (3) Wikipedia 1 Wikipedia Wikipedia 2 3 (1) (2) ( 2) nx R kl = α (tf sum S sum) + (tf i S i) (1) W kl = R kl / max(r km ) (2) tf sum k S sum k l tf i S i k l i=1

α Bowls Bowling greenbowling green Bowls Game tf i=2 S i=0.41 (1) 0.8 (2) 0 1 Bowling green 0.6 β 4. Wikipedia Wikipedia Wikipedia ( 3) Google Ajax API Microsoft Translate API Wikipedia (3) 0.

4 α Bowls Bowling greenbowling green Bowls Game tf i=2 S i=0.41 (1) 0.8 (2) 0 1 Bowling green 0.6 β 4. Wikipedia Wikipedia Wikipedia ( 3) Google Ajax API Microsoft Translate API Wikipedia (3) 0.3 P xi y i cos(x, y) = p P x 2 i P (3) yi 2 x 1 y 1 x i i y i 1 i 5. Wikipedia 3 Fig. 3 Compare contents [12] Wikipedia i = 1, 2,, M v i,j V V j = 0, 1,, N i j = 1 v i,1 i j = 0 v i,0 i Wikipedia e = 1, 2,, K e A e = {v i,j v i,j V and v i,j is written by e} j = 0 j v i,j 1 v i,j add i,j del i,j p (p = 0, 1,, N i j) v i,j+p add i,j del i,j p = 0 δ(add i,j, 0) = add i,j δ(del i,j, 0) = del i,j v i,j+p add i,j del i,j δ(add i,j, p), δ(del i,j, p) R add (i, j, p), R del (i, j, p) (4) (5) R add (i, j, p) = δ(add i,j, p) add i,j R del (i, j, p) = δ(del i,j, p) del i,j δ(add i,j, p), δ(del i,j, p), add i,j, del i,j δ(add i,j, p), δ(del i,j, p), add i,j, del i,j (4) (5)

5 p R add (i, j, p) R del (i, j, p) (6), (7) R add (i, j, p) = Radd (i, j, p) S add (p) R del (i, j, p) = Rdel (i, j, p) S del (p) S add (p), S del (p) p (6) (7) τ(v i,j ) (8) τ(v i,j) = N i j X q=1 N i j X R add (i, j, p) + q=1 R del (i, j, p) (8) 0 τ(v i,j) (9) τ(v i,j) = τ(v i,j) P M i=1 6. : α β P Ni j=1 τ(v i,j) P M i=1 Ni (9) α β α F Wikipedia Wikipedia α=7 10 F α=8 9 α=3 β=0.2 F α 3 β 0.2 F 2 Warwick Castle Warwick Castle Gaelic handball GAA Handball Gaelic handball 2 35% 57% 49% 59% F 1 # # 1 Bannock (food) 2 2 Warwick Castle 2 3 Black dog (ghost) 7 4 Fish and chips 4 5 Goodwood Festival of Speed 2 6 Bowls 2 7 Blue plaque 1 8 Burlesque 3 9 Flag of Scotland 6 10 Gaelic handball 4 11 Kipper 3 12 National Gallery of Scotland Lipton 1 Total 49 41% 58% Gaelic handball GAA Handball Gaelic handball S sum S i tf sum tf i F Gaelic handball Wikipedia Burlesque tf sum tf i Bowls Blue plaque Bowls Bowls World Bowls Events World Bowls Events Bowls World Bowls Events Blue plaque List of blue plaque (1) (2) 7.

6 4 :1 Fig. 4 Experiment:1 2 F (%) (%) F (%) (%) F Bannock (food) 0% Warwick Castle Black dog (ghost) Fish and chips Goodwood Festival of Speed Bowls Blue plaque Burlesque Flag of Scotland Gaelic handball Kipper National Gallery of Scotland Lipton Average Wikipedia Wikipedia Wikipedia [1] Michael Strube and Simone Paolo Ponzetto. Wikirelate! computing semantic relatedness using wikipedia. American Association for Artificial Intelligence, Vol. 2, pp , July [2] David Milne. Computing semantic relatedness using

7 wikipedia link structure. New Zealand Computer Science Research Student Conference, Vol. 7, p. 8, April [3] Kotaro Nakayama, Takahiro Hara, and Shojiro Nishio. Wikipedia mining for an association web thesaurus construction. Web Information Systems Engineering, Vol. 4831, pp , December [4] Yuku Takahashi, Hiroaki Ohshima, Mitsuo Yamamoto, Hirotoshi Iwasaki, Satoshi Oyama, and Katsumi Tanaka. Evaluating significance of historical entities based on tempospatial impacts analysis using wikipedia link structure. 22nd ACM conference on Hypertext and hypermedia, pp , June [5] Sergey Brin and Lawrence Page. The anatomy of a largescale hypertextual web search engine. 7th international conference on World Wide Web 7, pp , April [6] Jaap Kamps and Marijn Koolen. Is wikipedia link structure dierent? 2nd ACM International Conference on Web Search and Data Mining, pp , February [7] David Milne, Olena Medelyan, and Ian H. Witten. Mining domain-specific thesauri from wikipedia: A case study. The 2006 IEEE/WIC/ACM International Conference on Web Intelligence, pp , December [8] Zheng Chen, Shengping Liu, Liu Wenyin, Geguang Pu, and Wei-Ying Ma. Building a web thesaurus from web link structure. 26th annual international ACM SIGIR conference on Research and development in informaion retrieval, pp , July [9] Eytan Adar, Michael Skinner, and Daniel S. Weld. Information arbitrage across multi-lingual wikipedia. 2nd ACM International Conference on Web Search and Data Mining, pp , February [10] Thomas Wohner and Ralf Peters. Assessing the quality of wikipedia articles with lifecycle based metrics. 5th International Symposium on Wikis and Open Colaboration, pp. 1 10, October [11] Meiqun Hu. Measuring article quality in wikipedia: Models and evaluation. 16th ACM conference on Conference on information and knowledge management, pp , November [12],. Wikipedia. 21, pp ,

IPSJ SIG Technical Report Vol.2011-DBS-153 No /11/3 Wikipedia Wikipedia Wikipedia Extracting Difference Information from Multilingual Wiki

IPSJ SIG Technical Report Vol.2011-DBS-153 No /11/3 Wikipedia Wikipedia Wikipedia Extracting Difference Information from Multilingual Wiki Wikipedia 1 2 3 Wikipedia Wikipedia Extracting Difference Information from Multilingual Wikipedia Yuya Fujiwara, 1 Yu Suzuki 2 and Akiyo Nadamoto 3 There are multilingual articles on the Wikipedia. The