Enhanced anomaly detection using ensemble support vector machine

Identifying the intruder behavior in the network as well as in the system is an arduous and time-consuming mechanism. Abnormal activity of the user is a critical problem, analyzing this behavior with the emerging technologies like the Internet is difficulty. It should be investigated effectively and accurately for identification of all abnormal activities. The accurate anomaly detection is become a major problem in computer security. In the network environment data size is huge; identifying the abnormal activity from this huge data is the time consuming process. Detecting the anomaly from this data need more time, it is a critical problem in these days. The growth of computing speed is enabling us to detect it effectively; still there is a need to improve the anomaly detection time and accuracy. In this work, the rough set theory is used to reduce the dimensionality of the data to select the appropriate attributes in the detection process. Ensemble SVM is used to increase the detection accuracy and reduces the computation time effectively. The experimental results are proved.