本文關(guān)鍵詞： 機(jī)器學(xué)習(xí) 無監(jiān)督學(xué)習(xí) 聚類分析 學(xué)習(xí)分類系統(tǒng) 集成學(xué)習(xí) 復(fù)雜網(wǎng)絡(luò) 主干網(wǎng)提取　出處：《蘇州大學(xué)》2016年博士論文　論文類型：學(xué)位論文

【摘要】：無監(jiān)督學(xué)習(xí)是機(jī)器學(xué)習(xí)領(lǐng)域重要的研究方向之一,其應(yīng)用非常的廣泛。如數(shù)據(jù)聚類、復(fù)雜網(wǎng)絡(luò)的主干網(wǎng)提取等。本文以投票集成聚類和復(fù)雜網(wǎng)絡(luò)圖聚類為切入點(diǎn)進(jìn)行研究,取得的成績包括:(1)針對數(shù)據(jù)的集成聚類問題,提出了基于擴(kuò)展分類器系統(tǒng)的投票集成聚類方法。基于擴(kuò)展分類器系統(tǒng)的投票集成聚類方法,首先利用擴(kuò)展分類器系統(tǒng)在不同聚類個(gè)數(shù)的情況下生成一個(gè)聚類結(jié)果集合;然后引入分裂策略從所有候選值中確定聚類個(gè)數(shù);最后,采用基于多數(shù)投票的一致性方法獲得最終聚類結(jié)果。在人工數(shù)據(jù)和實(shí)際數(shù)據(jù)上的實(shí)驗(yàn)結(jié)果均表明了所提出方法的有效性。(2)在基于擴(kuò)展分類器系統(tǒng)的投票集成方法的基礎(chǔ)上,提出了基于擴(kuò)展分類器系統(tǒng)的統(tǒng)一聚類集成框架。該框架包括了更多適用的融合準(zhǔn)則、共識函數(shù)和自適應(yīng)集成等內(nèi)容。具體來說,在處理一個(gè)聚類任務(wù)的時(shí)候,所提出的方法首先會(huì)執(zhí)行學(xué)習(xí)分類器系統(tǒng)來生成幾個(gè)基聚類結(jié)果。為了使這些結(jié)果之間存在較大的多樣性,本文對聚類數(shù)據(jù)使用不同的初始化,比如使用不同的聚類數(shù)目等。得到這些基聚類結(jié)果之后,我們提出的方法會(huì)使用相應(yīng)的策略來生成最終的聚類結(jié)果。在人工數(shù)據(jù)和實(shí)際數(shù)據(jù)上的實(shí)驗(yàn)結(jié)果表明了所提框架的有效性。(3)針對復(fù)雜網(wǎng)絡(luò)的圖聚類問題,提出了一種基于不完全信息的無監(jiān)督學(xué)習(xí)的復(fù)雜網(wǎng)絡(luò)主干網(wǎng)提取方法。主干網(wǎng)提取的目的主要是壓縮復(fù)雜網(wǎng)絡(luò)的邊和點(diǎn)數(shù)量,以盡量精簡的子網(wǎng)絡(luò)保留原網(wǎng)絡(luò)的重要特征(如拓?fù)浣Y(jié)構(gòu)、點(diǎn)重要性特征等),從而幫助人們以更簡單的形式來理解網(wǎng)絡(luò)系統(tǒng)。本文以零模型為參考優(yōu)化邊過濾條件,并設(shè)計(jì)一種局部搜索模型。在四個(gè)真實(shí)網(wǎng)絡(luò)上的實(shí)驗(yàn)結(jié)果表明本文所提出方法不僅大幅度減少了主干網(wǎng)中的離群點(diǎn)、而且更好地保留了原網(wǎng)絡(luò)的各種特征、且比同類方法更加高效。
[Abstract]:Unsupervised learning is one of the important research directions in the field of machine learning. It is widely used, such as data clustering, backbone network extraction of complex networks, etc. The achievements of this paper include: (1) aiming at the problem of data clustering, an extended classifier system based voting ensemble clustering method is proposed, and an extended classifier system based voting ensemble clustering method is proposed. First, an extended classifier system is used to generate a set of clustering results under different clustering numbers; then a split strategy is introduced to determine the number of clusters from all candidate values. The final clustering results are obtained by using the consensus method based on majority voting. The experimental results on both artificial and actual data show that the proposed method is effective and based on the extended classifier system based voting integration method. A unified clustering integration framework based on extended classifier system is proposed. The framework includes more applicable fusion criteria, consensus functions and adaptive integration. The proposed method will first perform a learning classifier system to generate several base clustering results. In order to make these results more diverse, the clustering data are initialized differently in this paper. For example, using different clustering numbers, and so on. After obtaining these base clustering results, The proposed method will use the corresponding strategy to generate the final clustering results. The experimental results on artificial data and actual data show the effectiveness of the proposed framework. In this paper, an unsupervised learning method based on incomplete information is proposed to extract the backbone of complex networks. The main purpose of backbone network extraction is to compress the number of edges and points of complex networks. In order to help people understand the network system in a simpler form, the important features of the original network (such as topological structure, pointwise feature, etc.) are preserved by the subnetwork as concise as possible. In this paper, the zero model is used as the reference to optimize the edge filtering condition. Experimental results on four real networks show that the proposed method not only greatly reduces outliers in the backbone network, but also preserves all kinds of features of the original network. And more efficient than the same method.
【學(xué)位授予單位】：蘇州大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：TP181;O157.5

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 黃發(fā)良;黃名選;元昌安;姚志強(qiáng);;網(wǎng)絡(luò)重疊社區(qū)發(fā)現(xiàn)的譜聚類集成算法[J];控制與決策;2014年04期

，

本文編號：1526860