本文關(guān)鍵詞： 屬性圖聚類(lèi) 集群形成博弈 緊密性和均勻性約束 分布式學(xué)習(xí)算法 多智能體系統(tǒng)　出處：《計(jì)算機(jī)科學(xué)》2017年S1期 　論文類(lèi)型：期刊論文

【摘要】：近年來(lái)屬性圖聚類(lèi)受到了廣泛關(guān)注,其目的是將屬性圖中的節(jié)點(diǎn)劃分到若干簇中,使得每一個(gè)集群都有緊密的簇內(nèi)結(jié)構(gòu)和均勻的屬性值。現(xiàn)有的理論主要是假設(shè)屬性圖中的節(jié)點(diǎn)或?qū)ο笫菫榱藚f(xié)助優(yōu)化某個(gè)給定的方程,而忽略了它們?cè)诂F(xiàn)實(shí)生活中本身的屬性。同時(shí),一些開(kāi)放性問(wèn)題尚未得到有效解決,如異構(gòu)信息集成、計(jì)算成本高等。為此,把屬性圖聚類(lèi)問(wèn)題理解為自身節(jié)點(diǎn)代理的集群形成博弈。為了有效地整合拓?fù)浣Y(jié)構(gòu)和屬性信息,提出了基于緊密性和均勻性約束的節(jié)點(diǎn)代理策略選擇。進(jìn)一步證明了博弈過(guò)程將會(huì)收斂到弱帕累托納什均衡。在實(shí)證方面,設(shè)計(jì)了一個(gè)分布式和異構(gòu)的多智能體系統(tǒng),給出了一個(gè)快速的分布式學(xué)習(xí)算法。該算法的主要特點(diǎn)是結(jié)果分區(qū)的重疊率可以由一個(gè)事先給定的閾值控制。最后,在現(xiàn)實(shí)社交網(wǎng)絡(luò)上進(jìn)行了模擬實(shí)驗(yàn),并與目前先進(jìn)方法進(jìn)行比較,結(jié)果證實(shí)了所提算法的有效性。
[Abstract]:In recent years, attribute graph clustering has received extensive attention, the purpose of which is to divide the nodes in the attribute map into a number of clusters. So that each cluster has a tight cluster structure and uniform attribute value. The existing theory mainly assumes that the nodes or objects in the attribute map are to help optimize a given equation. At the same time, some open problems have not been solved effectively, such as heterogeneous information integration, high computing cost and so on. In order to integrate topology and attribute information effectively, the clustering problem of attribute graph is understood as the cluster game of its own node agent. The selection of node agent strategy based on compactness and uniformity constraints is proposed. It is further proved that the game process will converge to the weak Pareto Nash equilibrium. A distributed and heterogeneous multi-agent system is designed and a fast distributed learning algorithm is proposed. The main feature of the algorithm is that the overlap rate of the result partition can be controlled by a predetermined threshold. Finally. The simulation experiments on real social networks are carried out and compared with the current advanced methods. The results show that the proposed algorithm is effective.
【作者單位】： 中央財(cái)經(jīng)大學(xué)管理科學(xué)與工程學(xué)院;
【基金】：國(guó)家自然科學(xué)基金項(xiàng)目(71401194,71401188) 中央財(cái)經(jīng)大學(xué)“青年英才”培育支持項(xiàng)目(QYP1603)資助
【分類(lèi)號(hào)】：TP18;TP311.13
【正文快照】： 本文受?chē)?guó)家自然科學(xué)基金項(xiàng)目(71401194,71401188),中央財(cái)經(jīng)大學(xué)“青年英才”培育支持項(xiàng)目(QYP1603)資助。1引言許多現(xiàn)實(shí)中的信息系統(tǒng)是由大量高度關(guān)聯(lián)的參與者或?qū)ο蠼M成的,如在線社會(huì)網(wǎng)絡(luò)、無(wú)線傳感器網(wǎng)絡(luò)和眾包平臺(tái)。這些系統(tǒng)可以被屬性圖很好地模擬出來(lái),其中節(jié)點(diǎn)代表組件對(duì)，

本文編號(hào)：1466839