作者: Sangeeta Bhattacharya , S. Selvakumar
关键词: k-means clustering 、 Box plot 、 Cluster analysis 、 Flooding (computer networking) 、 Benchmark (computing) 、 Self-organizing map 、 Data mining 、 Computer science 、 Intrusion detection system
摘要: Multiconnection attacks such as DoS, probe, flooding, etc., have become common and attackers come out with sophisticated techniques well tools to launch variants of attacks. This growing amount attack sophistication has given rise the increasing need efficient detection algorithm. To test compare performances proposed algorithms, benchmark datasets are required represent dynamic nature network. Though certain available, most either synthetic or contains suppressed information. In this paper, we introduce SSENet-2014 dataset which is generated in a real network environment. The were using while carrying normal activities. description given. Then, comparison carried popular intrusion dataset, 10% KDD Cup 99. Two clustering approaches K Means Self Organizing Map (SOM) been used our experiments. Box plot analyze attributes two datasets. results confirm variability existing attribute values 99 dataset. Also, it can be seen that from varies considerably simulated traffic.