---

During the past few years, the number of malware designed for Android devices has increased dramatically. To confront with Android malware, some anomaly detection techniques have been proposed that are able to detect zero-day malware, but they often produce many false alarms that make them impractical for real-world use. In this paper, we address this problem by presenting DroidNMD, an ensemble-based anomaly detection technique that focuses on the network behavior of Android applications in order to detect Android malware. DroidNMD constructs an ensemble classifier consisting of multiple heterogeneous one-class classifiers and uses an ordered weighted averaging (OWA) operator to aggregate the outputs of the one-class classifiers. Our work is motivated by the observation that combining multiple one-class classifiers often produces higher overall classification accuracy than any individual one-class classifier. We demonstrate the effectiveness of DroidNMD using a real dataset of Android benign applications and malware samples. The results of our experiments show that DroidNMD can detect Android malware with a high detection rate and a relatively low false alarm rate.

---

This research investigates tool wear, elemental analysis (EDAX) on the machined surface, surface roughness, microhardness and microstructural changes in the cross-section of milled 304L stainless steel samples under dry and Minimum quantity lubrication (MQL) methods. The MQL process was able to improve the surface roughness for all milling parameters from 17% to 41% compared to the corresponding dry conditions. In dry machining, defects such as built up edge, severe flank wear and tool chipping were created. In MQL mode, these defects were significantly reduced and tool chipping was almost eliminated. By increase of cutting speed and depth, the surface hardness has increased. Compared to the dry method, the MQL reduces the hardness values and hardened depth below the machined surface. According to EDAX analysis on dry and MQL machined surfaces, applying the roughest cutting parameters, it was determined that no change of chemical elements occurred on machined surfaces. Increasing cutting parameters or dry machining causes the plastic deformation to intensify, the microstructure is flattened and the microstructure grains are compressed in the vicinity of the machined surface. The maximum reduction in thickness of deformed layer in MQL compared to dry method is 39%. For each milling sample, there is a direct relationship between the hardened depth and thickness of the corresponding microstructurally deformed layer.