Xingshu Chen

A Focused Crawler Based on Naive Bayes Classifier

The exponential growth of information on the World Wide Web makes it increasingly difficult to discover relevant data about a specific topic. In this case, growing interest is emerging in focused crawler, a program that traverses the Internet by choosing relevant pages to a predefined topic and neglecting those out of concern. A new focused crawler based on Naive Bayes classifier was proposed here, which used an improved TF-IDF algorithm to extract the characteristics of page content and adopted Bayes classifier to compute the page rank. Then the crawler developed was compared with a BFS crawler and a PageRank crawler, and the results show that our crawler has better performance than the PageRank crawler and BFS crawler in harvest ratio.

An Internet Traffic Identification Approach Based on GA and PSO-SVM

Internet traffic identification is currently an important challenge for network management. Many approaches have been proposed to classify different categories of Internet traffic. However, traditional approaches only focus on identifying TCP flows and have ignored the selection of best feature subset for classification. In this paper, we propose an approach to classify both TCP and UDP traffic flows using the Support Vector Machine (SVM) algorithm. In this approach, we select the best feature subset using Genetic Algorithm, and then we calculate the correspondence weight of each feature selected by Particle Swarm Optimization (PSO). In addition, the traditional SVM algorithm is optimized by PSO algorithm. The experimental results demonstrate that this approach can effectively select the feature subset from multiple attributes that can best reflect the differences among different network applications. Moreover, the identification rate is improved by the method of feature weighting and PSO optimized SVM algorithm.

Exploring the accuracy of capturing snapshots in large‐scale P2P IPTV systems

Summary With the growing maturity of peer-to-peer (P2P) technology, Internet protocol television (IPTV) applications based on that gained great success commercially and have attracted more and more attentions from both industry and academia. Currently, the active measurement method based on crawler technology is the most popular and effective one to study P2P IPTV systems. Existing measurement results revealed that accuracy of captured overlay snapshots depends on the crawling speed of crawler system. In order to capture more accurate overlay snapshots of P2P IPTV system, we developed a very fast and efficient distributed crawler system using the distributed architecture and peer degree-rank mechanism. In this paper, we first introduce the architectures of PPTV channel-list resource distribution and the whole system, which is the most popular and largest instance of P2P IPTV applications nowadays. Subsequently, this paper evaluates the crawling results of two dedicated crawlers capturing from peer-list servers and ordinary peers, respectively. Finally, we propose a fast and accurate dedicated crawler system based on distributed architecture and peer degree rank for PPTV. The experiment results show that the performance of our distributed crawler system is much better than other existing crawler systems. Specifically, our distributed crawler can track a very popular channel with about 7200 online users in 30 s. It is also reasonable to believe that our distributed crawler system can capture complete overlay snapshots. To the best of our knowledge, our study work is the first to explore capturing accurate overlay snapshots of large-scale P2P IPTV applications. Our crawler system can provide a good solution for capturing more accurate overlay snapshots of PPTV system and can also be used to help researchers to design crawler systems for other P2P IPTV systems. Copyright

An efficient method of P2P traffic identification based on wavelet packet decomposition and kernel principal component analysis

SUMMARY Peer-to-peer (P2P) traffic identification is currently an important challenge to network management and measurement. Many approaches based on statistics have been proposed to identify P2P traffic. However, flow features extracted by traditional methods are rough and one-sided, which might lead to inaccuracy identification of network traffic. Besides, P2P traffic has too many statistical features, which is a challenge to the time complexity and space complexity of the classifier. This work focuses on the study of flow features. First, micro features of flow signals are extracted based on wavelet packet decomposition, and we combine them with the traditional features into combination features. The experimental results show that combination features have better performance than traditional features for P2P traffic identification, and 16 kinds of wavelet functions were tested to find the best one. Second, a feature reduction algorithm based on improved kernel principal component analysis is provided. The results show that the feature reduction algorithm proposed in this paper plays good performance to P2P traffic identification, because it could greatly reduced the number of features while having no affection on identification accuracy. Copyright

IPTV-RM: A Resources Monitoring Architecture for P2P IPTV Systems

Resources monitoring is an important problem of the overall efficient usage and control of P2P IPTV systems. The resources of IPTV can include all distributing servers, programs and peers. Several researches have tried to address this issue, but most of them illuminated P2P traffic characterization, identification and user behavior. The main contributions of this paper are twofold. Firstly, a resources monitoring architecture for P2P IPTV systems, IPTV-RM, was presented based on previous work. The monitoring architecture employs a hierarchical structure and provides systemic monitoring including resources discovery, relative information extraction and analysis, trace and location. It gives a systematic framework for IPTV resources monitoring. Secondly, a distributed program crawling system (DMP-Crawler) was first proposed to collect information of programs, and a peer crawling system was put forward to harvest peers of a program. The results show that they are efficient and can be used for resource collection of other P2P system.

Measurement and Analysis of P2P IPTV Program Resource

With the rapid development of P2P technology, P2P IPTV applications have received more and more attention. And program resource distribution is very important to P2P IPTV applications. In order to collect IPTV program resources, a distributed multi-protocol crawler is proposed. And the crawler has collected more than 13 million pieces of information of IPTV programs from 2009 to 2012. In addition, the distribution of IPTV programs is independent and incompact, resulting in chaos of program names, which obstructs searching and organizing programs. Thus, we focus on characteristic analysis of program resources, including the distributions of length of program names, the entropy of the character types, and hierarchy depth of programs. These analyses reveal the disorderly naming conventions of P2P IPTV programs. The analysis results can help to purify and extract useful information from chaotic names for better retrieval and accelerate automatic sorting of program and establishment of IPTV repository. In order to represent popularity of programs and to predict user behavior and popularity of hot programs over a period, we also put forward an analytical model of hot programs.

A resources monitoring architecture for P2P file-sharing systems

Resources monitoring is an important problem of the overall efficient usage and control of P2P file-sharing systems. The resources of file-sharing systems can include all distributing servers, programs and peers. Several researches have tried to address this issue, but most of them illuminated P2P traffic characterization, identification and user behavior. Based on previous work, we present a resources monitoring architecture for P2P file-sharing systems. The monitoring architecture employs a hierarchical structure and provides systemic monitoring including resources discovery, relative information extraction and analysis, trace and location. It gives a systematic framework for file-sharing resources monitoring. And a prototype system has been developed based on the framework.