P2P traffic identification method based on an improvement incremental SVM learning algorithm

Abstract

How to classify the data sets with vast information amount and large distribution fluctuation, which is always the research hotspot. This paper puts forward an improved SVM incremental learning algorithm by comparing the different incremental learning methods of SVM algorithm. In the algorithm, whether to violate the KTT conditions is regarded as an important basis for incremental data set. And the algorithm will be more efficient on the classification of SVM incremental sets through optimizing and improving itself. Then the paper compares the SVM-based re-training algorithm, the standard SVM incremental learning algorithm and the improved SVM incremental learning algorithm through identifying P2P network traffic. The experimental results show that the improved SVM incremental learning algorithm proposed in this article can save storage space and increase the accuracy of the identification of P2P traffic.

Multi-Level P2P Traffic Classification Using Heuristic and Statistical-Based Techniques: A Hybrid Approach

“…However, it only classifies traffic associated with TCP flow and does not work on UDP flow. Gong et al [37] proposed an incremental algorithm to improve the learning of existing SVM, which has good space and time complexity and achieved the identification accuracy of 87.89% in identifying P2P traffic. Deng et al [38] proposed the ensemble learning model which uses the combination of random forests and feature weighted naive Bayes (FWNB) to classify P2P traffic.…”

Identifying P2P traffic: A survey

“…The performance achieved in terms of precision, false-positive and false-negative rates range from 96.55 to 97.89 %; 2 to 2.8 % and 2.45 to 5.29 %, respectively. Gong et al [89] proposed improved SVM incremental learning algorithm for P2P traffic identification which is able to save storage space and increase identification accuracy (87.89 %), when its performance is compared with standard SVM incremental learning algorithm (having 80.35 % accuracy) and SVM-based retraining algorithm (having 78.90 % accuracy) for increased number of test samples. Deng et al [90] proposed the ensemble learning model which integrates Random Forests and feature weighted Naive Bayes for P2P traffic identification.…”