Published Papers - KAZAMA Kazuhiro

語の共有に基づく文書ネットワークの構造的特徴について

佐藤 進也, 風間 一洋, 福田 健介, 村上 健一郎

情報処理学会論文誌   47 ( 3 ) 829 - 839   2006.03  [Refereed]

トラックバックネットワークに基づくSEOコミュニティの分析

風間 一洋, 佐藤 進也, 斉藤 和巳, 木村 昌弘

情報処理学会論文誌   47 ( 3 ) 840 - 849   2006.03  [Refereed]

Network analyses to understand the structure of wikipedia

Takeshi Yamada, Kazumi Saito, Kazuhiro Kazama

Proceedings of AISB'06: Adaptation in Artificial and Biological Systems   3   195 - 198   2006

　View Summary

We investigated the structure of Wikipedia, the well known Web-based encyclopedia, as a very largescale hyperlink network by using network analysis methods. We analyzed the time evolution of the number of articles and links, applied core-extraction methods to identify highly interconnected subnetworks, and compared different centrality measures to understand its characteristics.

Evaluation of using human relationships on the web as information navigation paths

Kazuhiro Kazama, Shin-ya Sato, Kensuke Fukuda, Ken-ichiro Murakami, Hiroshi Kawakami, Osamu Katai

NEW FRONTIERS IN ARTIFICIAL INTELLIGENCE ( SPRINGER-VERLAG BERLIN )  4012 ( 4012 ) 302 - 313   2006  [Refereed]

　View Summary

We investigated the use of human relationships on the web for information navigation paths. We propose a new information navigation method that uses personal names. It automatically extracts the human relationships of key people by analyzing the co-occurences of personal names on a web page from search results that are relevant to a specific topic and provides two facilities for using these relationships as information navigation paths. One is information navigation using a list of the key people and a list of related people. Another is information navigation using a network diagram of the key people. We consider human. relationships on the web as new information navigation paths like hyperlinks. We analyzed the network structure of human relationships for various topics and evaluated their usefulness in order to clarify the applicable scope and improve the usefulness. The results show that human relationships are adequate shortcut networks for search results for most cases. However, if the ratio of the number of personal names to the number of web pages is too high, the relationships of the people are too tight for information navigation and our future work is to reduce the number of edges without reducing the coverage of search results. If the degree of density of human-related information in the higher ranked search results is too low, human activity is low and our method is not suitable.

DOI

Extracting communities from complex networks by the K-dense method

Kazumi Saito, Takeshi Yamada, Kazuhiro Kazama

ICDM 2006: Sixth IEEE International Conference on Data Mining, Workshops ( IEEE COMPUTER SOC )    300 - 304   2006  [Refereed]

　View Summary

To understand the structural and functional properties of large-scale complex networks, it is crucial to efficiently extract a set of cohesive subnetworks as communities. There have been proposed several such community extraction methods in the literature, including the classical k-core decomposition method and, more recently, the k-clique based community extraction method. The k-core method, although computationally efficient, is often not powerful enough for uncovering a detailed community structure and it produces only coarse-grained and loosely connected communities. The k-clique method, on the other hand, can extract fine-grained and tightly connected communities but requires a substantial amount of computational load for large-scale complex networks. In this paper, we present a new notion of a subnetwork called k-dense, and propose an efficient algorithm for extracting k-dense communities. We applied our method to the two different types of networks assembled from real data, namely, from blog trackbacks and word associations, demonstrated that the k-dense method could extract communities almost as efficiently as the k-core method, while the qualities of the extracted communities are comparable to those obtained by the k-clique method.

Web空間における人間関係を用いた情報探索の一手法

風間 一洋, 佐藤 進也, 福田 健介, 村上 健一郎, 川上 浩司, 片井 修

情報処理学会論文誌：データベース   46 ( SIG 13 (TOD27) ) 26 - 39   2005.09  [Refereed]

実世界指向Webマイニングによる同姓同名人物の分離

佐藤 進也, 風間 一洋, 福田 健介, 村上 健一郎

情報処理学会論文誌：データベース   46 ( SIG 8 (TOD26) ) 26 - 36   2005.06  [Refereed]

分散協調翻訳支援システムの構築と運用

小山 博史, 風間 一洋, 岡本 隆史, 横田 健彦

情報処理学会論文誌   46 ( 4 ) 974 - 984   2005.04  [Refereed]

実世界指向Webマイニングの提案とその同姓同名人物分離問題への適用

佐藤 進也, 風間 一洋, 福田 健介, 村上 健一郎

DBSJ Letters ( 日本データベース学会 )  3 ( 4 ) 21 - 24   2005.03  [Refereed]

Detecting Search Engine Spam from a Trackback Network in Blogspace

Masahiro Kimura, Kazumi Saito, Kazuhiro Kazama, Shin-ya Sato

KES 2005: Knowledge-Based Intelligent Information and Engineering Systems, LNCS ( Springer Berlin Heidelberg )  3684   723 - 729   2005  [Refereed]

　View Summary

We aim to develop a technique to detect search engine optimization (SEO) spam websites. Specifically, we propose four methods for extracting the SEO spam entries from a given trackback network in blogspace that are based on fundamental metrics on a network. Using real data of trackback networks in blogspace, we experimentally evaluate the performance of the proposed methods, and demonstrate that the method of ranking entries based on average degrees of nearest neighbors can be a very promising approach for extracting SEO spam entries.

DOI

Autonomous development of relationships among web servers for facilitating information utilization

Shin-Ya Sato, Masanori Harada, Kazuhiro Kazama

Systems and Computers in Japan   35 ( 10 ) 40 - 49   2004.09  [Refereed]

　View Summary

A Web server can play a role as an agent for people who create information, and Web servers with related information can be searched by the agent in order to facilitate efficient information utilization on the Web. Web servers exchange information with other servers and select which ones to keep or abandon, resulting in an increase in the links with servers with greater relevance. As a result, a community, a group of Web servers that are mutually linked, is created. A Web server community can be used as a place for various forms of information utilization and exchange, such as meta-searches and contents syndication. In this paper, the authors propose an algorithm for creating communities after evaluating the meaning of facilitating information utilization via a community. They then verify the validity of their idea through simulations. © 2004 Wiley Periodicals, Inc.

DOI

Webディレクトリ拡張の自動化手法

風間 一洋, 原田 昌紀, 佐藤 進也

情報処理学会論文誌：データベース   45 ( SIG 7 (TOD22) ) 218 - 229   2004.06  [Refereed]

Query Networkによる情報発見・収集支援

佐藤 進也, 風間 一洋, 原田 昌紀

情報処理学会論文誌：データベース   45 ( SIG 7 (TOD22) ) 1 - 10   2004.06  [Refereed]

Query Networkの構造と時間発展を利用した情報発見・収集支援

佐藤 進也, 原田 昌紀, 風間 一洋

DBSJ Letters ( 日本データベース学会 )  2 ( 4 ) 11 - 14   2004.03  [Refereed]

Finding authoritative people from the Web

Masanori Harada, Shin-Ya Sato, Kazuhiro Kazama

Proceedings of the ACM IEEE International Conference on Digital Libraries, JCDL 2004 ( Association for Computing Machinery (ACM) )    306 - 313   2004

　View Summary

Today's web is so huge and diverse that it arguably reflects the real world. For this reason, searching the web is a promising approach to find things in the real world. This paper presents NEXAS, an extension to web search engines that attempts to find real-world entities relevant to a topic. Its basic idea is to extract proper names from the web pages retrieved for the topic. A main advantage of this approach is that users can query any topic and learn about relevant real-world entities without dedicated databases for the topic. In particular, we focus on an application for finding authoritative people from the web. This application is practically important because once personal names are obtained, they can lead users from the web to managed information stored in digital libraries. To explore effective ways of finding people, we first examine the distribution of Japanese personal names by analyzing about 50 million Japanese web pages. We observe that personal names appear frequently on the web, but the distribution is highly influenced by automatically generated texts. To remedy the bias and find widely acknowledged people accurately, we utilize the number of web servers containing a name instead of the number of web pages. We show its effectiveness by an experiment covering a wide range of topics. Finally, we demonstrate several examples and suggest possible applications.

DOI

Autonomous Development of Relationships among Wed Servers for Facilitating Information Utilization

SATO Shin-ya, HARADA Masanori, KAZAMA Kazuhiro

The Transactions of the Institute of Electronics,Information and Communication Engineers. ( The Institute of Electronics, Information and Communication Engineers )  J86-D-I ( 8 ) 583 - 591   2003.08  [Refereed]

　View Summary

Web上の効率的な情報利用を支援するため,Webサーバに情報発信者のエージェントとしての役割を与え,関連情報をもつWebサーバを探索させる. Webサーバは,互いに他のサーバに関する情報を交換し,取捨選択することで,より関連性の高いサーバとのつながりを増やしていく.その結果,相互に関連するWebサーバの集団であるコミュニティが形成される.Webサーバのコミュニティは,メタサーチやコンテンツシンジケーションなど種々の情報利用・流通の場として利用できる.本論文では,まず,コミュニティによる情報利用支援の意義を検討した上で,コミュニティ形成のアルゴリズムを提案し,シミュレーションによってその妥当性を検証する.

サーチエンジンの検索結果のマルチレベルグルーピングの評価

風間 一洋, 原田 昌紀, 佐藤 進也

コンピュータソフトウェア ( 日本ソフトウェア科学会 )  17 ( 4 ) 58 - 69   2000.12  [Refereed]

Evaluation of Multi-level Grouping Scheme for the Search Results that A Search Engine Generates.

KAZAMA Kazuhiro, HARADA Masanori, SATO Shin-ya, Kazuhiro Kazama, Masanori Harada, Shin - ya Sato, NTT Network Innovation Laboratories., NTT Network Innovation Laboratories., NTT Network Innovation Laboratories.

Computer Software ( Japan Society for Software Science and Technology )  17 ( 4 ) 354 - 365   2000.07

DOI

WWWトラフィック解析による情報生産者と情報消費者の数量的特徴付け

佐藤 進也, 風間 一洋, 清水 奨, 神林 隆

情報処理学会論文誌   40 ( 7 ) 2851 - 2860   1999.07  [Refereed]

HTML Document Correlation Analysis by User's Behavior in World Wide Web Navigation

KAZAMA Kazuhiro, SATO Shin-ya, SHIMIZU Susumu, KAMBAYASHI Takashi

Transactions of Information Processing Society of Japan ( Information Processing Society of Japan (IPSJ) )  40 ( 5 ) 2450 - 2459   1999.05  [Refereed]

　View Summary

This paper describes the method to construct HTML document correlation graph that indicates mutual relations of them by analysis of user's access sequences to them and their URL 2-grams, and defines the correlation by the graph structure and path's frequency. And we evaluate the results that were acquired by this method in the small community.