Faculty Profiles - KAZAMA Kazuhiro

Published Papers - KAZAMA Kazuhiro

Division display61 - 80 of about 100／ All the affair displays >>

情報拡散影響度に基づく機能コミュニティ抽出法

伏見卓恭, 斉藤和巳, 池田哲夫, 風間一洋

人工知能学会全国大会論文集 ( 一般社団法人人工知能学会 ) 2013 ( 0 ) 2J15 - 2J15 2013

<p>従来の機能コミュニティ抽出法では，ネットワーク上でのランダムウォークを仮定したPageRank計算において，各ステップでの確率ベクトルの収束過程を基にノードの機能を定義し，類似機能を有するノード群を抽出していた．本稿では，ネットワーク上での情報拡散に着目し，他のノードへの影響力を基にノードの機能を定義した機能コミュニティ概念，および抽出手法を提案する．</p>

DOI

ソーシャルネットワークによるWebからの情報収集

風間一洋

情報の科学と技術 63 ( 1 ) 28 - 33 2013 [Invited]

個々のノードの視点に基づく機能コミュニティ抽出法

伏見卓恭, 斉藤和巳, 池田哲夫, 風間一洋, 松尾真人

WebDB Forum 2012 ( 情報処理学会データベースシステム研究会など ) 2012.11 [Refereed]

Twitter上のリプライ，リツイート，ハッシュタグの類似構造分析

小出明弘, 斉藤和巳, 風間一洋, 鳥海不二夫

WebDB Forum 2012 ( 情報処理学会データベースシステム研究会など ) 2012.11 [Refereed]

Characteristics Analysis of Multi Graphs Focused on Co-link Structure

小出明弘, 斉藤和巳, 風間一洋, 鳥海不二夫

日本データベース学会論文誌 ( 日本データベース学会 ) 11 ( 1 ) 13 - 18 2012.10 [Refereed]

機能性に基づくコミュニティ抽出法の比較

伏見卓恭, 斉藤和巳, 風間一洋

情報処理学会論文誌：データベース 5 ( 3 ) 26 - 35 2012.09 [Refereed]

ネットワーク機能コミュニティ抽出法

伏見卓恭, 斉藤和巳, 風間一洋

日本データベース学会論文誌 10 ( 3 ) 13 - 18 2012.02 [Refereed]

Characteristics of information diffusion in blogs, in relation to information source type

Kazuhiro Kazama, Miyuki Imada, Keiichiro Kashiwagi

Neurocomputing ( Elsevier BV ) 76 ( 1 ) 84 - 92 2012.01 [Refereed]

A novel method is presented to analyze the dynamics of social media, i.e., information diffusion properties, for information recommendation and ranking. In social media such as blogs, various information diffuses over time. As a result, a network structure is constructed. In an information diffusion network, each influential information source has an affected subnetwork whose nodes are reachable from it. We define three information diffusion properties of the subnetwork using the numbers of three types of directed two-edge connected subgraphs, which are basic structures in a directed acyclic graph such as an information diffusion network. Each basic structure type is related to information scattering, information gathering, or information transmission. We visualized and analyzed the structure of information diffusion networks extracted for various topics. Furthermore, we characterized the information diffusion properties by using the rank correlation coefficient, precision, and mean reciprocal rank and mean average precision of three types of information sources: official sites, news articles, and consumer generated media pages. We found that the three information diffusion properties have different characteristics and give priority to different types of information sources. © 2011 Elsevier B.V.

DOI

Extracting Communities in Networks Based on Functional Properties of Nodes.

Takayasu Fushimi, Kazumi Saito, Kazuhiro Kazama

Knowledge Management and Acquisition for Intelligent Systems - 12th Pacific Rim Knowledge Acquisition Workshop(PKAW) ( Springer ) 328 - 334 2012

DOI

Studies on Relationship between TV Reports and Twitter Information Concerning the Fukushima Dai-ichi Nuclear Power Plants Accident:(2)Analysis of Word Co-occurrence Structure Concerning Radioactivity

SATO Ryosuke, KIMURA Hiroshi, TORIUMI Fujio, SAKAKI Takeshi, KAZAMA Kazuhiro, FUKUDA Kensuke

Proceedings of Annual / Fall Meetings of Atomic Energy Society of Japan ( Atomic Energy Society of Japan ) 2012 8 - 8 2012

本研究のシリーズ報告である「(1)テレビ報道からツイッターへの情報伝播に関する分析」によって、テレビ報道とツイッターの間に関連性があることが示唆された。そこで本研究は、放射能を中心とした単語を含む情報に注目し、ツイッターとテレビ報道での放射能に関連する用語の共起構造を比較することで、ツイッターとテレビ報道との関連性を明らかにする。

DOI

Distribution pattern analysis of associated geographical names on transportation network

Kazuhiro Kazama

Transactions of the Japanese Society for Artificial Intelligence ( Japanese Society for Artificial Intelligence ) 27 ( 2 ) 34 - 39 2012 [Refereed]

I present a new text mining approach combined with network analysis to quantify the distribution patterns of the associated geographical names on a transportation network. I extract geographical names from user's search queries recorded in search engine query logs and compare the similarities of any pair of geographical names using Jaccard coefficient. I found that a set of associated geographical names for each geographical name shows a specific spatial distribution pattern on transportation network and define a measure for quantifying such characteristics. Furthermore, I discuss its characteristics and application for information navigation.

DOI

Studies on Relationship between TV Reports and Twitter Information Concerning the Fukushima Dai-ichi Nuclear Power Plants Accident:(1) Analysis of Information Propagation from TV Reports to Twitter

KIMURA Hiroshi, SATO Ryosuke, SHIBATA Yogo, TORIUMI Fujio, SAKAKI Takeshi, KAZAMA Kazuhiro, FUKUDA Kensuke

Proceedings of Annual / Fall Meetings of Atomic Energy Society of Japan ( Atomic Energy Society of Japan ) 2012 7 - 7 2012

2011年3月11日に起こった東日本大震災、それに引き続いて起こった東京電力・福島第一原子力発電所事故（以降、福島原発事故）は、インターネットを含むさまざまなメディアによって扱われ、社会に大きな影響を与えたといわれる。本研究では、代表的なメディアのひとつであるテレビ報道と、インターネット上のソーシャルメディアのひとつであるツイッターとに着目し、福島原発事故に関連する情報がそれぞれのメディアでどのように扱われたのか、また、それらの情報にはどのような関連性があったのかを分析する。なお、本報告は「（２）放射能に関連する用語についての共起構造分析」とのシリーズ報告である。

DOI

情報伝播特性の定量化手法

風間一洋, 今田美幸, 柏木啓一郎

コンピュータソフトウェア 28 ( 1 ) 162 - 172 2011.02 [Refereed]

Quantification method of information diffusion properties

Kazuhiro Kazama, Miyuki Imada, Keiichiro Kashiwagi

Computer Software 28 ( 1 ) 162 - 172 2011

This paper proposes a quantification method of three information diffusion properties of the subnetwork that is reachable from an information source in the information diffusion network. This subnetwork is a directed acyclic graph and is composed of three types of directed 2-edge connected subgraphs, which is related to the basic phenomena of information diffusion: information scattering, information gathering, and information transmitting. We define the information scatter degree, the information gather degree, and information transmit degree as information diffusion properties by using the number of three types of directed 2-edge connected subgraphs. We analyze the characteristics and the time series variation of information diffusion attributes in real information diffusion networks extracted from blogspace by these information diffusion properties.

Characteristics estimation of information sources by information diffusion analysis

Kazuhiro Kazama, Miyuki Imada, Keiichiro Kashiwagi

Proceedings - 2010 IEEE/WIC/ACM International Conference on Web Intelligence, WI 2010 1 484 - 491 2010 [Refereed]

This paper presents a novel method to estimate characteristics of information sources about a topic by analyzing their information diffusion subnetworks in blogspace. In an information diffusion network, each influential information source has an affected subnetwork whose nodes are reachable from it. We define three information diffusion properties of the subnetwork using the numbers of three types of directed 2-edge connected subgraphs, which are basic structures in a directed acyclic graph such as an information diffusion network. Each type of basic structure is related to information scattering, information gathering, or information transmission. We visualized and analyzed the structure of information diffusion networks extracted for other topics. Furthermore, we characterize the information diffusion properties by using the rank correlation coefficient, precision, and mean reciprocal rank (MRR) and mean average precision (MAP) of three types of information sources: official sites, news sites, and CGMs. We found that the three information diffusion properties have different characteristics and give priority to different types of information sources. © 2010 IEEE.

DOI

Quantification of information diffusion network characteristics in blogspace

Kazuhiro Kazama, Miyuki Imada, Keiichiro Kashiwagi

Transactions of the Japanese Society for Artificial Intelligence ( Japanese Society for Artificial Intelligence ) 25 ( 3 ) 404 - 409 2010 [Refereed]

This paper proposes three new quantification measures of information diffusion characteristics in blogspace. These measures are calculated from the number of three basic structures, which are directed 2-edge connected sub graphs, in information diffusion networks. Each basic structure is related to information scattering, information gathering or information transmission. We analyze and visualize information diffusion networks extracted from six blog datasets. In the result, we show that the difference of information diffusion characteristics can be discriminated by the combination of three measures and human activity in blogspace can be explained by them.

DOI

The k-dense method to extract communities from complex networks

Kazumi Saito, Takeshi Yamada, Kazuhiro Kazama

Studies in Computational Intelligence 165 243 - 257 2009 [Refereed]

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 only discovers 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 three different types of networks assembled from real data, namely, from blog trackbacks, word associations and Wikipedia references, and 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. © 2009 Springer-Verlag Berlin Heidelberg.

DOI

Extracting Communities from Complex Networks by the k-Dense Method

Kazumi Saito, Takeshi Yamada, Kazuhiro Kazama

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES ( IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG ) E91A ( 11 ) 3304 - 3311 2008.11 [Refereed]

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 three different types of networks assembled from real data, namely, from blog trackbacks, word associations and Wikipedia references, and 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.

DOI

Extraction of Overlapping Community Structures from Human Relationships.

風間一洋, 佐藤進也, 斉藤和巳, 山田武士

コンピュータソフトウェア 24 ( 1 ) 81 - 90 2007.06 [Refereed]

リアルタイム圧縮によるパケットキャプチャの高速化

清水奨, 風間一洋, 廣津登志夫, 後藤滋樹

情報学会論文誌：コンピューティングシステム 47 ( SIG 7(ACS14) ) 183 - 193 2006.05 [Refereed]

　 PREV - NEXT 　

<KAZAMA Kazuhiro>

Published Papers - KAZAMA Kazuhiro