NNS:A Novel Neighborhood Negative Selection algorithm

As the security issue becomes more complex, more and more anomaly detection schemes involve high-dimension data. Negative selection algorithms have been widely used in anomaly detection, fault detection, and fraud detection. However, these algorithms perform poorly when dealing with high- dimension data. To address this issue, we propose a novel Neighborhood Negative Selection (NNS) algorithm in this paper. In NNS, we use a neighborhood set to represent a self-sample (or a detector), instead of a single data point. As a result, the delay for training detectors is greatly reduced. We further introduce a special matching mechanism to limit the negative effect of the dimensionality of a shape space and improve the detecting performance in high dimensions. The experimental results show that NNS can provide a more accurate and stable detection performance. Meanwhile, both theoretical analysis and experimental results show that NNS further improves the training efficiency.