網(wǎng)絡(luò)流量特征選擇方法中的分治投票策略研究
發(fā)布時間:2018-10-07 17:24
【摘要】:特征選擇作為機器學(xué)習(xí)過程中的預(yù)處理步驟,是影響分類性能的關(guān)鍵因素.網(wǎng)絡(luò)流量具有數(shù)據(jù)量大,特征維度高的特點,如何快速提取特征子集,并提高分類效率對于基于機器學(xué)習(xí)的流量分類方法具有重要意義.本文提出基于分治與投票策略的特征提取方法,將數(shù)據(jù)集分裂為多個子集,分別執(zhí)行特征提取算法,利用投票方法獲得最后的特征子集.實驗表明可有效提高特征提取的時間效率,同時使分類器取得良好的分類準(zhǔn)確率.
[Abstract]:Feature selection, as a preprocessing step in machine learning, is a key factor affecting classification performance. Network traffic is characterized by large amount of data and high feature dimension. How to quickly extract feature subsets and improve classification efficiency is of great significance to the traffic classification method based on machine learning. In this paper, a feature extraction method based on divide-conquer and voting strategy is proposed. The data set is divided into multiple subsets, and the feature extraction algorithm is implemented separately. The last feature subset is obtained by voting method. Experiments show that it can effectively improve the time efficiency of feature extraction and make the classifier achieve good classification accuracy.
【作者單位】: 浙江大學(xué)計算機學(xué)院;浙江科技學(xué)院理學(xué)院;浙江大學(xué)圖書與信息中心;嘉興職業(yè)技術(shù)學(xué)院;
【基金】:國家973重點基礎(chǔ)研究發(fā)展計劃(No.2012CB315903) 浙江省重點科技創(chuàng)新團(tuán)隊(No.2011R50010-21,No.2013TD20) 國家自然科學(xué)基金(No.61379118) 國家科技支撐計劃(No.2014BAH24F01) 國家863計劃(No.2012AA01A507) 浙江省網(wǎng)絡(luò)媒體云處理與分析工程技術(shù)中心開放課題(No.2012E10023-14)
【分類號】:TP393.06
[Abstract]:Feature selection, as a preprocessing step in machine learning, is a key factor affecting classification performance. Network traffic is characterized by large amount of data and high feature dimension. How to quickly extract feature subsets and improve classification efficiency is of great significance to the traffic classification method based on machine learning. In this paper, a feature extraction method based on divide-conquer and voting strategy is proposed. The data set is divided into multiple subsets, and the feature extraction algorithm is implemented separately. The last feature subset is obtained by voting method. Experiments show that it can effectively improve the time efficiency of feature extraction and make the classifier achieve good classification accuracy.
【作者單位】: 浙江大學(xué)計算機學(xué)院;浙江科技學(xué)院理學(xué)院;浙江大學(xué)圖書與信息中心;嘉興職業(yè)技術(shù)學(xué)院;
【基金】:國家973重點基礎(chǔ)研究發(fā)展計劃(No.2012CB315903) 浙江省重點科技創(chuàng)新團(tuán)隊(No.2011R50010-21,No.2013TD20) 國家自然科學(xué)基金(No.61379118) 國家科技支撐計劃(No.2014BAH24F01) 國家863計劃(No.2012AA01A507) 浙江省網(wǎng)絡(luò)媒體云處理與分析工程技術(shù)中心開放課題(No.2012E10023-14)
【分類號】:TP393.06
【參考文獻(xiàn)】
相關(guān)期刊論文 前1條
1 寧卓;孫知信;龔儉;張維維;;利用流量特征的GIDS報文分類優(yōu)化算法[J];電子學(xué)報;2012年03期
【共引文獻(xiàn)】
相關(guān)期刊論文 前10條
1 李斌;;網(wǎng)絡(luò)流量分類及其現(xiàn)狀研究[J];廣西教育;2013年39期
2 周亞建;薛超;平源;;基于端口特征的P2P應(yīng)用識別方案[J];北京工業(yè)大學(xué)學(xué)報;2013年11期
3 李為民;劉曉楠;繆晨;陳陸穎;雷振明;;典型業(yè)務(wù)的包長分布規(guī)律[J];電子科技大學(xué)學(xué)報;2014年02期
4 錢亞冠;張e,
本文編號:2255036
本文鏈接:http://sikaile.net/guanlilunwen/ydhl/2255036.html
最近更新
教材專著