本文選題：流量分類(TC)　切入點：ReliefF　出處：《高技術(shù)通訊》2016年02期

【摘要】：研究了被廣泛應(yīng)用于互聯(lián)網(wǎng)流量分類的樸素貝葉斯分類方法的性能特點,針對此方法在給定類別下給出的所有流量特征同等重要并且是獨立的假設(shè)在現(xiàn)實中難以滿足,致使分類準確率不高的問題,提出一種基于特征加權(quán)的樸素貝葉斯流量分類算法。該算法基于NetFlow記錄的特征信息,采用特征選擇算法ReliefF和相關(guān)系數(shù)方法計算每個特征的權(quán)重值,然后將網(wǎng)絡(luò)流量分配至后驗概率最大的應(yīng)用類別中。實驗結(jié)果表明,這種基于特征加權(quán)的樸素貝葉斯算法具有超過94%的分類準確率,并且維持了樸素貝葉斯方法簡單高效、分類穩(wěn)定的特性,可以滿足當前高帶寬網(wǎng)絡(luò)流量分類的需求。
[Abstract]:This paper studies the performance characteristics of naive Bayes classification method, which is widely used in Internet traffic classification. In view of the assumption that all traffic characteristics given by this method under a given class are equally important and independent, it is difficult to satisfy in reality. Because the accuracy of classification is not high, a naive Bayesian traffic classification algorithm based on feature weighting is proposed. Based on the feature information recorded by NetFlow, the feature selection algorithm ReliefF and the correlation coefficient method are used to calculate the weights of each feature. Then the network traffic is allocated to the application category with the largest posterior probability. The experimental results show that the feature weighted naive Bayesian algorithm has more than 94% classification accuracy and maintains the simplicity and efficiency of the naive Bayesian method. Classification stability can meet the needs of current high bandwidth network traffic classification.
【作者單位】： 中國科學(xué)院計算機網(wǎng)絡(luò)信息中心;中國科學(xué)院大學(xué);
【基金】：973計劃(2012CB315803) 中國科學(xué)院計算機網(wǎng)絡(luò)信息中心“一三五”計劃(CNIC_PY-1401)資助項目
【分類號】：TP393.06

【相似文獻】

相關(guān)期刊論文 前4條

1 崔中杰;姚淑萍;唐成華;;基于特征加權(quán)的故障征兆處理算法[J];計算機工程與應(yīng)用;2006年34期

2 崔燕;汪斌強;陳庶樵;張震;;基于行為特征加權(quán)的P2P流識別方法的研究[J];計算機工程與設(shè)計;2009年20期

3 胡亮;任維武;任斐;劉曉博;金剛;;基于改進密度聚類的異常檢測算法[J];吉林大學(xué)學(xué)報(理學(xué)版);2009年05期

4 ;[J];;年期

，

本文編號：1676813