Biblio

Network behavior anomaly detection (NBAD) require fast mechanisms for learning from the large scale data. However, the training velocity of general machine learning approach is largely limited by the adopted training weights of all features in the NBAD. In this paper, we notice, however, that the related weights matching of NBAD features is sparse, which is not necessary for holding all weights. Hence, in this paper, we consider an efficient support vector machine (SVM) approach for NBAD by imposing 1 -norm. Essentially, we propose to use sparse SVM (S-SVM), where sparsity in model, i.e. in weights is used to interfere with special feature selection and that can achieve feature selection and classification efficiently.

Hexagonal CdS and Fe-doped CdS nanorods were synthesised by a facile hydrothermal method and characterised by X-ray diffraction, energy dispersive X-ray spectroscopy, UV-vis absorption, photoluminescence, and X-ray photoelectron spectroscopy. The magnetic properties of undoped and Fe-doped CdS nanorods were investigated at room temperature. The experimental results demonstrate that the ferromagnetism of the Fe-doped CdS nanorods differs from that of the undoped CdS nanorods. The remanence magnetisation (Mr) and the coercive field (Hc) of the Fe-doped CdS nanorods were 4.9 × 10-3 emu/g and 270.6 Oe, respectively, while photoluminescence properties were not influenced by doping. First-principle calculations show that the ferromagnetism in Fe-doped CdS nanocrystal arose not only from the Fe dopants but also from the Cd vacancies, although the main contribution was due to the Fe dopants.