现代计算机(专业版)
現代計算機(專業版)
현대계산궤(전업판)
MODERN COMPUTER
2014年
10期
3-6
,共4页
P2P网络流量识别%支持向量机%增量学习%过渡分类超平面
P2P網絡流量識彆%支持嚮量機%增量學習%過渡分類超平麵
P2P망락류량식별%지지향량궤%증량학습%과도분류초평면
Network Traffic Identification%SVM(Support Vector Machine)%Incremental Learning%Temporary Classification Hyperplane
针对标准支持向量机在P2P网络流量识别中不支持增量学习的问题,提出一种适于P2P网络流量识别的SVM快速增量学习方法。在对违背Karush-Kuhn-Tucker条件的新增正负样本集分别进行聚类分析基础上,运用聚类簇中心对支持向量机训练生成一个接近增量学习最优分类超平面的过渡超平面,并以此超平面为基准确定初始训练样本集上非支持向量和支持向量的互相转化,进而生成新的样本集实现SVM增量学习。理论分析和实验结果表明,该方法能有效简化增量学习的训练样本集,在不降低P2P网络流量识别精度的前提下,明显缩短SVM的增量学习时间和识别时间。
針對標準支持嚮量機在P2P網絡流量識彆中不支持增量學習的問題,提齣一種適于P2P網絡流量識彆的SVM快速增量學習方法。在對違揹Karush-Kuhn-Tucker條件的新增正負樣本集分彆進行聚類分析基礎上,運用聚類簇中心對支持嚮量機訓練生成一箇接近增量學習最優分類超平麵的過渡超平麵,併以此超平麵為基準確定初始訓練樣本集上非支持嚮量和支持嚮量的互相轉化,進而生成新的樣本集實現SVM增量學習。理論分析和實驗結果錶明,該方法能有效簡化增量學習的訓練樣本集,在不降低P2P網絡流量識彆精度的前提下,明顯縮短SVM的增量學習時間和識彆時間。
침대표준지지향량궤재P2P망락류량식별중불지지증량학습적문제,제출일충괄우P2P망락류량식별적SVM쾌속증량학습방법。재대위배Karush-Kuhn-Tucker조건적신증정부양본집분별진행취류분석기출상,운용취류족중심대지지향량궤훈련생성일개접근증량학습최우분류초평면적과도초평면,병이차초평면위기준학정초시훈련양본집상비지지향량화지지향량적호상전화,진이생성신적양본집실현SVM증량학습。이론분석화실험결과표명,해방법능유효간화증량학습적훈련양본집,재불강저P2P망락류량식별정도적전제하,명현축단SVM적증량학습시간화식별시간。
In P2P network traffic identification, aims to such the problems that SVM does not support incremental learning. Proposes a fast incremen-tal learning method of SVM for P2P network traffic identification. After clustering of positive and negative training samples that violate Karush-Kuhn-Tucker conditions, a temporary classification hyperplane close to classification hyperplane of incremental learning is ob-tained by using clustering centers to train standard SVM. Based on it, transfers support-vector and non-support-vector in original training samples to produce new training samples for incremental learning of SVM. Analysis and simulation shows that the method effectively sim-plifies training samples of incremental learning and greatly reduces the training and traffic identification time of SVM in incremental learning.