局部加權(quán)單類支持向量機(jī)研究
發(fā)布時間：2018-09-08 08:19
【摘要】：單類分類是介于監(jiān)督學(xué)習(xí)和無監(jiān)督學(xué)習(xí)之間的機(jī)器學(xué)習(xí)任務(wù),它能夠有效地解決僅有一類樣本訓(xùn)練分類器的問題和類別極端不平衡問題。迄今為止,涌現(xiàn)了大量的單類分類方法,其中單類支持向量機(jī)(one-class support vector machine,OCSVM)是最為常用的方法之一。然而,傳統(tǒng)的單類支持向量機(jī)存在一些不足,如:沒有考慮到訓(xùn)練樣本的幾何分布對其分類器性能的影響�；诖�,本文利用訓(xùn)練樣本的局部幾何信息,對單類支持向量機(jī)從間隔改進(jìn)和錯分樣本加權(quán)兩方面開展了研究。1.提出了基于局部相關(guān)保留的單類支持向量機(jī)(locality correlation preserving based one-class support vector machine,LCP-OCSVM)。所提方法將局部相關(guān)保留(locality correlation preserving,LCP)與單類支持向量機(jī)相結(jié)合,繼承了LCP與OCSVM的優(yōu)點,可在保留樣本的局部相關(guān)性的同時最大化特征空間中樣本的像與原點之間的間隔。在人工數(shù)據(jù)集及標(biāo)準(zhǔn)數(shù)據(jù)集上驗證了所提方法的可行性。2.提出了局部保留加權(quán)的單類支持向量機(jī)(locality preserving weighted one-class support vector machine,LPWOCSVM)。為了降低錯分樣本對單類支持向量機(jī)分類邊界的影響,所提方法根據(jù)訓(xùn)練樣本的局部幾何信息構(gòu)造局部保留加權(quán)向量,為錯分樣本分配較小權(quán)重,使得單類支持向量機(jī)的分類邊界更為緊致。在人工數(shù)據(jù)集以及標(biāo)準(zhǔn)數(shù)據(jù)集上的實驗表明,所提方法具有更強(qiáng)的抗噪聲能力和更優(yōu)的泛化性能。
[Abstract]:Single class classification is a machine learning task between supervised learning and unsupervised learning. It can effectively solve the problem of only one kind of sample training classifier and extreme class imbalance problem. Up to now, a large number of single class classification methods have emerged, among which single class support vector machine (one-class support vector machine,OCSVM) is one of the most commonly used methods. However, there are some shortcomings in traditional single-class SVM, such as not considering the influence of geometric distribution of training samples on the performance of classifier. Based on this, this paper makes use of the local geometric information of the training samples to study the single-class support vector machine from the two aspects of interval improvement and misdivision sample weighting. A single class support vector machine (locality correlation preserving based one-class support vector machine,LCP-OCSVM) based on local correlation reservation is proposed. The proposed method combines local correlation preserving (locality correlation preserving,LCP) with single class support vector machine, inherits the advantages of LCP and OCSVM, and maximizes the interval between the image and origin of the sample in the feature space while preserving the local correlation of the sample. The feasibility of the proposed method is verified on the human data set and the standard data set. 2. 2. A locally reserved weighted single class support vector machine (locality preserving weighted one-class support vector machine,LPWOCSVM) is proposed. In order to reduce the influence of misdivided samples on the classification boundary of single class support vector machines, the proposed method constructs locally reserved weighted vectors according to the local geometric information of the training samples, and assigns a smaller weight to the misdivided samples. The classification boundary of single class support vector machine is more compact. Experiments on artificial data sets and standard datasets show that the proposed method has better anti-noise capability and better generalization performance.
【學(xué)位授予單位】：河北大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP181
本文編號：2229947
[Abstract]:Single class classification is a machine learning task between supervised learning and unsupervised learning. It can effectively solve the problem of only one kind of sample training classifier and extreme class imbalance problem. Up to now, a large number of single class classification methods have emerged, among which single class support vector machine (one-class support vector machine,OCSVM) is one of the most commonly used methods. However, there are some shortcomings in traditional single-class SVM, such as not considering the influence of geometric distribution of training samples on the performance of classifier. Based on this, this paper makes use of the local geometric information of the training samples to study the single-class support vector machine from the two aspects of interval improvement and misdivision sample weighting. A single class support vector machine (locality correlation preserving based one-class support vector machine,LCP-OCSVM) based on local correlation reservation is proposed. The proposed method combines local correlation preserving (locality correlation preserving,LCP) with single class support vector machine, inherits the advantages of LCP and OCSVM, and maximizes the interval between the image and origin of the sample in the feature space while preserving the local correlation of the sample. The feasibility of the proposed method is verified on the human data set and the standard data set. 2. 2. A locally reserved weighted single class support vector machine (locality preserving weighted one-class support vector machine,LPWOCSVM) is proposed. In order to reduce the influence of misdivided samples on the classification boundary of single class support vector machines, the proposed method constructs locally reserved weighted vectors according to the local geometric information of the training samples, and assigns a smaller weight to the misdivided samples. The classification boundary of single class support vector machine is more compact. Experiments on artificial data sets and standard datasets show that the proposed method has better anti-noise capability and better generalization performance.
【學(xué)位授予單位】：河北大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP181
【參考文獻(xiàn)】
相關(guān)期刊論文 前3條
1 皋軍;王士同;鄧趙紅;;基于全局和局部保持的半監(jiān)督支持向量機(jī)[J];電子學(xué)報;2010年07期
2 馮愛民;劉學(xué)軍;孫廷凱;;嵌入數(shù)據(jù)結(jié)構(gòu)信息的單類支持向量機(jī)及其線性規(guī)劃算法[J];中山大學(xué)學(xué)報(自然科學(xué)版);2009年06期
3 賈銀山,賈傳熒;一種加權(quán)支持向量機(jī)分類算法[J];計算機(jī)工程;2005年12期
，本文編號：2229947
本文鏈接：http://sikaile.net/kejilunwen/zidonghuakongzhilunwen/2229947.html
最近更新
教材專著
