1 Broder Navigational URL URL Informational Web Transactional Web Web Web 2 Broder [16] SearchLife Broder [16] Daniel [17] Broder

Size: px

Start display at page:

Download "1 Broder Navigational URL URL Informational Web Transactional Web Web Web 2 Broder [16] SearchLife Broder [16] Daniel [17] Broder"

しなつめいこ
5 years ago
Views:

1 DEIM Forum 2010 B gs08062@s.inf.shizuoka.ac.jp, {yokoyama,fukuta,ishikawa}@inf.shizuoka.ac.jp Web Web. Evaluation of a Multiple Viewpoints Clustering Search Engine Based on a Classification Model of Search Queries Ryosuke MURAMATSU, Shohei YOKOYAMA, Naoki FUKUTA, and Hiroshi ISHIKAWA Graduate School of Informatics, Shizuoka University Johoku, Naka-ku, Hamamatsu-shi, Shizuoka, Japan Department of Computer Science, Faculty of Informatics, Shizuoka University Johoku, Naka-ku, Hamamatsu-shi, Shizuoka, Japan gs08062@s.inf.shizuoka.ac.jp, {yokoyama,fukuta,ishikawa}@inf.shizuoka.ac.jp Abstract Web search engines play important roles as a tool to obtain useful information from the web. Since keyword-based search engines produce a long list of results that contain links to relevant web pages, approaches for improving visibility of results to grasp an overview of the search results has investigated. In this paper, we propose a system that clusterizes search results with multiple cluster labels for improved view of results. To make labels be a useful guide, we provide a multi way labeling method that considers statistical features of words in documents. We evaluate our approaches based on a search query classification model, and show our labeling methods for clustering search system have advantages in some complex searching tasks. Key words infomation retrieval, data mining 1. Web 1 Google [1] Yahoo [2] [3] [15].

2 1 Broder Navigational URL URL Informational Web Transactional Web Web Web 2 Broder [16] SearchLife Broder [16] Daniel [17] Broder Navigational Informational Transactional 3 Broder 1 Daniel Broder Informational Transactional Resource Daniel Resource Web 2. 2 Web 1 Web 1 Web Zamir [18] [8] Zamir Suffix Tree STC METAL Web Clusty [5] Carrot 2 [14] Clusty Web Clusty Velocity remix remix remix Web Carrot 2 Carrot 2 Lingo [13] STC [6] Lingo Imafuji [19] [7] Imafuji Web Web

3 ブラウザ SearchLife 外部 Web サービス入力画面出力画面 html 検索クエリ ( タイトル, サマリ URL) DB タイトル情報提示部タイトル 1 取得部検索クエリ処理部クラスタリング部ラベル情報 idf, 特徴語検索クエリラベル構築部 (XML) 名詞ヒット件数名詞クラスタ情報検索クエリ (http リクエスト ) Yahoo! Web 検索 API Yahoo! 形態素解析 API 2. 3 Zamir [6] STC Grouper Zamir Web Songhua [4] google Broder [16] 2. 4 Web Web ( ) Jae-wook [20] [21] Jae-wook Web Jae-wook Web 3 2 Web Web Web Web Web 3 4 Web 4 3 Web Web Web 1 Web Web [22] [23] [9] [23]

1 ( 1 ( 1 ) Carrot 2 [14]( 2 ) ( 3 ) 4 2 [13], [15]

4 Web 4 Web [23] Broder [16] Informational ( 1 ( 1 ) Carrot 2 [14]( 2 ) ( 3 ) 4 2 [13], [15] , Lingo Yahoo! 3 Yahoo! [24] Web

3 1( ) 1.1 2010 2010 3 1.2 1.3 3 1.4 2 4 2( ) 2.

5 3 1( ) ( ) ( ) URL Web

6 1 1 2 3 4 3 7 3 3 1 6 4 4 2 2 0 0 7 9 3 6 4 4 2 6 Carrot 2 7 2 1 2 3 4 7 4 3 2 1 0 6 3 7 2 5 6 3 2 3 4 2

1 10 1 2 1( ) 0.54 0.63 0.15 0.62 2( ) 0.67 0.31 0.63 0.44 3( ) 0.72 0.71 0.27 0.

6 Carrot Web Web Web 2 3 Web ( ) ( ) ( ) (1) 1 A support(a) TotalPoint(A) 1 A,Examinee(A) 1 A support(a) = 4. 2 T otalp oint(a) 3 Examinee(A) (1)

7 9 ( ) 1( 1) 2(Carrot) 3() ( ) ( ) ( ) ( ) , Web

8 Web SearchLife 1 B( ) [1] Googlehttp:// [2] Yahoo! [3] P. Ferragina, and A. Gulli A personalized search engine based on web-snippet hierarchical clustering WWW 05: Special interest tracks and posters of the 14th international conference on World Wide Web pp ACM [4] S. Xu, T. Jin, and F.C. Lau, A new visual search interface for web browsing, Proceedings of the Second ACM International Conference on Web Search and Data Mining, pp , ACM, [5] Clustyhttp:// [6] O. Zamir, and O. Etzioni, Grouper: A dynamic clustering interface to web search results, WWW 99: Proceedings of the 8th international World Wide Web Conference, pp , Elsevier North-Holland, Inc., [7] Max flow web (DBSJ Letters) vol.4 no.2 pp [8] Technical report DE [9] 19 B [10] M.A. Hearst, and J.O. Pedersen, Reexamining the cluster hypothesis:scater/gather on retrieval results, SIGIR 96, pp.76 84, ACM, [11] P. Pantel, and D. Lin, Document clustering with committees, SIGIR 02, pp , ACM, [12] Web 14 2-p [13] S. Osinski J. Stefanowski, and D. Weiss Lingo:search results clustering algorithm based on singular value decomposition Proceedings of the International IIS: IIPWM 04 Conference pp [14] Carrot 2 [15] J. Stefanowski, and D. Weiss., Carrot 2 and language properties in web search results clustering, In Proceedings of AWIC-2003, First International Atlantic Web Intelligence Conference, vol.2663 of Lecture Notes in Computer Science, pp , Springer, [16] A. Broder, A taxonomy of web search, SIGIR Forum 36, vol.36, pp.3 10, ACM, [17] D.E. Rose, and D. Levinson, Understanding user goals in web search, WWW 03: Proceedings of the 13th international conference on World Wide Web, pp.13 19, ACM, [18] O. Zamir, and O. Etzioni, Web document clustering : a feasibility demonstration, Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval, [19] N. Imafuji, and M. Kitsuregawa, Finding web communities by maximum flow algorithm using well-assigned edge capacities, IEICE transactions on information and systems, vol.e87-d(2), pp , [20] J.W. Ahn, P. Brusilovsky, D. He, J. Grady, and Q. Li, Personalized web exploration with task models, Proceeding of the 17th international conference on World Wide Web, pp.1 10, ACM, [21] (DBSJ Letters) vol.6 no.2 pp [22] 146 idb 2008 pp [23] [24] Yahoo! co.jp/.

4_6.dvi

4_6.dvi Vol. 50 No. 4 1399 1409 (Apr. 2009) 1, 1 1 1 1 TSUBAKI 1 Web Information Observation Using Keyword Distillation Based Clustering Yasuo Bamba, 1, 1 Keiji Shinzato, 1 Tomohide Shibata 1 and Sadao Kurohashi