本文選題：半監(jiān)督　切入點(diǎn)：特征選擇　出處：《東南大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：本學(xué)位論文的工作基于國(guó)家電網(wǎng)公司科技項(xiàng)目"電力信息通信網(wǎng)絡(luò)流量預(yù)測(cè)和管道智能化關(guān)鍵技術(shù)研究及其應(yīng)用"項(xiàng)目,以面向業(yè)務(wù)的流量識(shí)別與感知為研究目標(biāo),主要內(nèi)容為基于半監(jiān)督學(xué)習(xí)的網(wǎng)絡(luò)業(yè)務(wù)流量識(shí)別方法研究。針對(duì)網(wǎng)絡(luò)流量具有較多特征屬性的特點(diǎn),通過(guò)研究特征選擇算法,提出了一種基于信息度量的半監(jiān)督特征選擇算法(IMSFS,Information Measure based Semi-supervised Feature Selection);針對(duì) DBSCAN(Density-Based Spatial Clustering of Applications with Noise)算法難以確定輸入?yún)?shù)和數(shù)據(jù)集含有大量重復(fù)數(shù)據(jù)等問(wèn)題,提出了一種改進(jìn)的DBSCAN算法;還將IMSFS算法與改進(jìn)的DBSCAN算法結(jié)合,提出了一種半監(jiān)督DBSCAN(SDBSCAN,Semi-supervised DBSCAN)流量識(shí)別方法。在 Moore 數(shù)據(jù)集上的驗(yàn)證結(jié)果表明,本文提出的SDBSCAN流量識(shí)別方法在包含少量已標(biāo)記樣本的流量集中具有較高的分類準(zhǔn)確率。全文共分為五章,各章節(jié)的主要內(nèi)容為:第一章概述課題研究背景及研究目的,分析了流量識(shí)別的發(fā)展趨勢(shì)和需要解決的關(guān)鍵問(wèn)題,同時(shí)給出本論文的章節(jié)安排。第二章概述了流量識(shí)別,分析了半監(jiān)督學(xué)習(xí)依賴的假設(shè)和半監(jiān)督學(xué)習(xí)的分類等基本知識(shí),為后續(xù)的研究奠定基礎(chǔ)。第三章針對(duì)網(wǎng)絡(luò)流量具有較多特征屬性的特點(diǎn),通過(guò)研究特征選擇算法,提出了一種基于信息度量的半監(jiān)督特征選擇算法(IMSFS)并在Moore數(shù)據(jù)集上進(jìn)行了驗(yàn)證。第四章針對(duì)DBSCAN算法難以確定輸入?yún)?shù)和數(shù)據(jù)集含有大量重復(fù)數(shù)據(jù)等問(wèn)題,提出了一種改進(jìn)的DBSCAN算法;還將IMSFS算法與改進(jìn)的DBSCAN算法結(jié)合,提出了一種半監(jiān)督DBSCAN流量識(shí)別方法并在Moore數(shù)據(jù)集上進(jìn)行了驗(yàn)證。第五章總結(jié)本學(xué)位論文的研究工作,并指出進(jìn)一步研究方向。
[Abstract]:The work of this dissertation is based on the project of State Grid Corporation "Research and Application of key Technologies for Traffic Prediction and Pipeline Intelligence in Power Information and Communication Networks", which aims at traffic identification and perception oriented to business. The main content is the research of network traffic identification method based on semi-supervised learning. In this paper, a semi-supervised feature selection algorithm based on information metric is proposed, and an improved DBSCAN algorithm is proposed to solve the problem that DBSCAN(Density-Based Spatial Clustering of Applications with Noisealgorithm is difficult to determine the input parameters and that the data set contains a large number of repetitive data. By combining the IMSFS algorithm with the improved DBSCAN algorithm, a semi-supervised DBSCANN semi-supervised DBSCANs traffic identification method is proposed. The verification results on the Moore dataset show that, The SDBSCAN traffic identification method proposed in this paper has a high classification accuracy in traffic concentration containing a small number of labeled samples. The whole paper is divided into five chapters. The main contents of each chapter are as follows: chapter 1 summarizes the background and purpose of the research. The development trend of traffic identification and the key problems to be solved are analyzed, and the chapter arrangement of this paper is given. In chapter 2, the basic knowledge of traffic identification is summarized, and the hypothesis of semi-supervised learning dependence and the classification of semi-supervised learning are analyzed. The third chapter aims at the characteristics of network traffic with more feature attributes, and studies the feature selection algorithm. A semi-supervised feature selection algorithm based on information metric is proposed and validated on the Moore dataset. Chapter 4th aims at the problem that the DBSCAN algorithm is difficult to determine the input parameters and the data set contains a large number of repeated data. An improved DBSCAN algorithm is proposed, and a semi-supervised DBSCAN traffic identification method is proposed by combining the IMSFS algorithm with the improved DBSCAN algorithm, which is validated on the Moore dataset. Chapter 5th summarizes the research work of this dissertation. The further research direction is pointed out.
【學(xué)位授予單位】：東南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TM73

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 姜羽;華俊;胡靜;宋鐵成;劉世棟;郭經(jīng)紅;;電力信息通信網(wǎng)絡(luò)流量識(shí)別技術(shù)研究[J];信息化研究;2015年01期

2 劉建偉;劉媛;羅雄麟;;半監(jiān)督學(xué)習(xí)方法[J];計(jì)算機(jī)學(xué)報(bào);2015年08期

3 湯憶則;馬平;高鈞利;張磊;項(xiàng)肖峰;;基于智能電網(wǎng)的一體化信息通信網(wǎng)絡(luò)發(fā)展研究[J];數(shù)字通信;2014年04期

4 李雙慶;慕升弟;;一種改進(jìn)的DBSCAN算法及其應(yīng)用[J];計(jì)算機(jī)工程與應(yīng)用;2014年08期

5 黎運(yùn)發(fā);黃名輝;;核密度估計(jì)逐點(diǎn)最優(yōu)窗寬選擇的改進(jìn)[J];統(tǒng)計(jì)與決策;2011年14期

6 顧海林;申揚(yáng);;業(yè)務(wù)識(shí)別與流量管理在電力信息化中的應(yīng)用[J];電力信息化;2010年12期

7 尤鳴宇;陳燕;李國(guó)正;;不均衡問(wèn)題中的特征選擇新算法:Im-IG[J];山東大學(xué)學(xué)報(bào)(工學(xué)版);2010年05期

8 趙國(guó)鋒;吉朝明;徐川;;Internet流量識(shí)別技術(shù)研究[J];小型微型計(jì)算機(jī)系統(tǒng);2010年08期

9 徐鵬;林森;;基于C4.5決策樹(shù)的流量分類方法[J];軟件學(xué)報(bào);2009年10期

10 夏魯寧;荊繼武;;SA-DBSCAN:一種自適應(yīng)基于密度聚類算法[J];中國(guó)科學(xué)院研究生院學(xué)報(bào);2009年04期

，

本文編號(hào)：1643964