【摘要】：在傳統(tǒng)的動(dòng)態(tài)社團(tuán)探測(cè)方法中,由于每個(gè)時(shí)間片網(wǎng)絡(luò)之間相互獨(dú)立,無(wú)法高效地探測(cè)社團(tuán)并分析社團(tuán)的演化事件。針對(duì)傳統(tǒng)方法的不足,提出一種利用時(shí)序網(wǎng)絡(luò)的歷史信息,解決上述兩個(gè)問(wèn)題。對(duì)于每個(gè)時(shí)間片網(wǎng)絡(luò),僅計(jì)算連邊發(fā)生改變的節(jié)點(diǎn);根據(jù)社團(tuán)的定義及上一時(shí)刻的社團(tuán)信息,探測(cè)當(dāng)前時(shí)刻的社團(tuán)并分析它們的演化事件。在人工網(wǎng)絡(luò)上的實(shí)驗(yàn)結(jié)果表明,相對(duì)傳統(tǒng)方法,該方法能夠保證社團(tuán)劃分的質(zhì)量并分析社團(tuán)的演化事件,提升了探測(cè)效率。
[Abstract]:In the traditional dynamic community detection method, because each time slice network is independent of each other, it is unable to detect the community and analyze the evolution events of the community efficiently. In view of the shortcomings of traditional methods, this paper presents a method to solve the above two problems by using the historical information of time series network. For each time slice network, only the nodes whose edges have changed are calculated. According to the definition of the community and the information of the community at the last moment, the community at the current time is detected and their evolution events are analyzed. The experimental results on artificial networks show that this method can guarantee the quality of community division and analyze the evolution events of communities, and improve the detection efficiency.
【作者單位】： 電子科技大學(xué)互聯(lián)網(wǎng)中心;電子科技大學(xué)大數(shù)據(jù)研究中心;
【基金】：國(guó)家自然科學(xué)基金項(xiàng)目(61433014、61673085) 中央高校基本科研基金項(xiàng)目(ZYGX2014Z002)
【分類(lèi)號(hào)】：O157.5

【相似文獻(xiàn)】

相關(guān)期刊論文 前1條

1 王建;馮偉森;邱興超;劉繼;盧林;;基于BP模型的KAD網(wǎng)絡(luò)核心節(jié)點(diǎn)識(shí)別算法研究[J];計(jì)算機(jī)工程與應(yīng)用;2013年07期

相關(guān)碩士學(xué)位論文 前1條

1 王偉;基于核心節(jié)點(diǎn)的局部社團(tuán)發(fā)現(xiàn)及相關(guān)研究[D];華東理工大學(xué);2014年

，

本文編號(hào)：2133313