本文選題：演化聚類　切入點：時間平滑性　出處：《中國科學技術(shù)大學》2017年碩士論文　論文類型：學位論文

【摘要】：隨著互聯(lián)網(wǎng)的蓬勃發(fā)展和人們采集數(shù)據(jù)能力的增強,實際生活中出現(xiàn)了越來越多的隨時間變化的數(shù)據(jù),我們稱這類數(shù)據(jù)為演化數(shù)據(jù)(EvolutionaryData)。近年來,演化數(shù)據(jù)的聚類問題引起了很多研究者的興趣。一般地,演化聚類的要求有兩個方面:1)每個時刻發(fā)現(xiàn)的聚類結(jié)構(gòu)要盡可能好地劃分當前時刻的快照數(shù)據(jù);2)每個時刻發(fā)現(xiàn)的聚類結(jié)構(gòu)要盡可能保持時間平滑性,即和上個時刻相比,當前時刻發(fā)現(xiàn)的聚類結(jié)構(gòu)盡量不發(fā)生太大的變化。演化數(shù)據(jù)聚類有很廣泛的應(yīng)用背景,其研究有著很重要的意義。本文從核心節(jié)點和參考點的角度來研究演化聚類。本文的主要內(nèi)容包括三個方面。1)受靜態(tài)社區(qū)發(fā)現(xiàn)算法Top Leaders啟發(fā),我們提出一個基于核心節(jié)點(LeaderNodes)的演化社區(qū)發(fā)現(xiàn)算法(EvoLeaders)。首先,我們基于結(jié)合時間信息的更新策略來得到每個時刻的初始核心節(jié)點。通過保持發(fā)現(xiàn)的初始核心節(jié)點集合與上個時刻核心節(jié)點集合的時間平滑性,來保證由這些核心節(jié)點發(fā)現(xiàn)的社區(qū)跟以前的結(jié)構(gòu)盡量保持平滑。然后,通過一組分裂合并操作提高社區(qū)質(zhì)量。在兩個實際數(shù)據(jù)集上的實驗結(jié)果表明,EvoLeaders算法比Top Leaders算法效果更好。該工作表明了從核心節(jié)點的角度進行演化社區(qū)發(fā)現(xiàn)的可行性。2)Top Leaders算法的主要缺點是需要人工輸入社區(qū)數(shù)目。基于網(wǎng)絡(luò)中每個節(jié)點與其鄰居節(jié)點之間度的關(guān)系,以及節(jié)點之間共同鄰居的重疊程度,我們改進了 Top Leaders算法,并提出了能夠自動發(fā)現(xiàn)社區(qū)數(shù)目的AutoLeaders算法。在三個經(jīng)典數(shù)據(jù)集上的實驗結(jié)果表明,AutoLeaders算法不僅能夠發(fā)現(xiàn)合理的社區(qū)數(shù)目,還能夠發(fā)現(xiàn)合理的社區(qū)結(jié)構(gòu)。進一步,基于兩種時間平滑性策略,我們提出了在動態(tài)網(wǎng)絡(luò)中發(fā)現(xiàn)社區(qū)的新的解決方案,即EvoAutoLeaders算法。在兩個實際數(shù)據(jù)集上的結(jié)果表明EvoAutoLeaders算法的效果比較好。3)我們從參考點的角度來處理演化聚類問題。首先,我們引入了三種不同的參考點,以及相應(yīng)的計算個體到參考點距離的策略。然后,基于r-dominance關(guān)系和多目標演化算法,提出了一個演化聚類算法(即rEvoC算法)。實驗結(jié)果證明,與經(jīng)典算法相比,rEvoC算法更適合聚類演化數(shù)據(jù),而且能夠取得更好的效果�？偟膩碚f,我們從核心節(jié)點和參考點的角度來處理演化數(shù)據(jù)聚類問題,并且通過實驗證明了其有效性,而且效果比經(jīng)典算法更優(yōu)。本文的工作對演化社區(qū)發(fā)現(xiàn)和演化數(shù)據(jù)聚類方法研究方面具有一定的參考價值。
[Abstract]:With the rapid development of the Internet and the enhancement of people's ability to collect data, more and more data have changed over time in real life, which we call evolutionary data in recent years. The clustering of evolutionary data has aroused the interest of many researchers. The evolutionary clustering requirement has two aspects: 1) the cluster structure discovered at each moment should be as well divided as possible into the snapshot data of the current moment. (2) the clustering structure discovered at each moment should be as smooth as possible, that is, compared with the previous moment. The clustering structure discovered at present is as little as possible. Evolutionary data clustering has a wide range of applications. This paper studies evolutionary clustering from the point of view of core nodes and reference points. The main contents of this paper include three aspects. 1) inspired by the static community discovery algorithm Top Leaders. We propose an evolutionary community discovery algorithm based on core node LeaderNodes.First, We obtain the initial core nodes at each moment based on the update strategy combined with time information, by maintaining the temporal smoothness of the initial core node set and the core node set at the last moment. To ensure that the community found by these core nodes remains as smooth as possible from the previous structure. Then, The experimental results on two real data sets show that the EvoLeaders algorithm is more effective than the Top Leaders algorithm. This work shows that the EvoLeaders algorithm can be found in the evolutionary community from the point of view of the core nodes. The main drawback of the row. 2n Leaders algorithm is the need to manually input the number of communities. Based on the degree relationship between each node in the network and its neighbor node, And the degree of overlap of the common neighbors between nodes, we improve the Top Leaders algorithm, The experimental results on three classical data sets show that the AutoLeaders algorithm can not only find the reasonable number of communities, but also find the reasonable community structure. Based on two time smoothing strategies, we propose a new solution for community discovery in dynamic networks. That is, EvoAutoLeaders algorithm. The results on two actual data sets show that the effect of EvoAutoLeaders algorithm is better. 3) We deal with the evolutionary clustering problem from the point of view of reference points. First, we introduce three different reference points. Then, based on r-dominance relation and multi-objective evolutionary algorithm, an evolutionary clustering algorithm (i.e. rEvoC algorithm) is proposed. Compared with the classical algorithm, the rEvoC algorithm is more suitable for clustering evolutionary data and can achieve better results. In general, we deal with the problem of evolutionary data clustering from the point of view of core nodes and reference points, and the experimental results show that the algorithm is effective. And the result is better than the classical algorithm. The work of this paper has some reference value for the research of evolutionary community discovery and evolutionary data clustering method.
【學位授予單位】：中國科學技術(shù)大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP311.13

【相似文獻】

相關(guān)期刊論文 前10條

1 劉秀燕;消除參考點接近失敗報警方法[J];機床電器;2002年01期

2 劉韜;;數(shù)控車床回參考點與回參考點的故障排除[J];機電產(chǎn)品開發(fā)與創(chuàng)新;2009年01期

3 劉建峰;;關(guān)于西門子810系統(tǒng)機床返回參考點的故障分析[J];機械工人.冷加工;2006年01期

4 李可心;;關(guān)于機床回參考點的維修與診斷探究[J];科技傳播;2012年11期

5 樊留群,羅建軍,姜迪剛;機床參考點的研究[J];機械制造;1997年06期

6 劉瑞已;現(xiàn)代數(shù)控機床回參考點的故障分析與排除[J];現(xiàn)代制造工程;2005年03期

7 劉瑞已;數(shù)控機床回參考點故障分析與診斷[J];設(shè)備管理與維修;2005年01期

8 李晗靜;李生;趙鐵軍;;漢語中方位參考點恢復(fù)研究[J];計算機研究與發(fā)展;2007年02期

9 胡慶;王龍;;3G終端中S參考點的實現(xiàn)研究[J];電信科學;2007年03期

10 鄭宇;;加工中心回參考點的特殊故障及處理[J];制造技術(shù)與機床;2007年04期

相關(guān)會議論文 前7條

1 李剛斌;;數(shù)控機床參考點的設(shè)置與維修[A];第八屆全國設(shè)備與維修工程學術(shù)會議、第十三屆全國設(shè)備監(jiān)測與診斷學術(shù)會議論文集[C];2008年

2 黃萬長;;帶距離編碼標記的線性測量系統(tǒng)返參考點[A];第三屆數(shù)控機床與自動化技術(shù)專家論壇論文集[C];2012年

3 徐平;;西門子840D系統(tǒng)伺服軸參考點調(diào)整方法研究[A];2011年“天山重工杯”全國機電企業(yè)工藝年會暨第五屆機械工業(yè)節(jié)能減排工藝技術(shù)研討會論文集[C];2011年

4 王春來;;數(shù)控機床回參考點報警類故障及實例分析[A];綠色制造與低碳經(jīng)濟——2010年海南省機械工程學會、海南省機械工業(yè)質(zhì)量管理協(xié)會“年會”暨機械工程科技學術(shù)報告會論文集[C];2010年

5 王春來;;數(shù)控機床回參考點報警類故障及實例分析[A];“綠色制造 質(zhì)量管理”——海南省機械工程學會、海南省機械工業(yè)質(zhì)量管理協(xié)會2011年會論文集[C];2011年

6 李震;江億;劉曉華;陳曉陽;;濕空氣熱力過程的，

本文編號：1599880