Malicious Code Variant Identification Based on Multiscale Feature Fusion CNNs

The increasing volume and types of malwares bring a great threat to network security. The malware binary detection with deep convolutional neural networks (CNNs) has been proved to be an effective method. However, the existing malware classification methods based on CNNs are unsatisfactory to this day because of their poor extraction ability, insufficient accuracy of malware classification, and high cost of detection time. To solve these problems, a novel approach, namely, multiscale feature fusion convolutional neural networks (MFFCs), was proposed to achieve an effective classification of malware based on malware visualization utilizing deep learning, which can defend against malware variants and confusing malwares. The approach firstly converts malware code binaries into grayscale images, and then, these images will be normalized in size by utilizing the MFFC model to identify malware families. Comparative experiments were carried out to verify the performance of the proposed method. The results indicate that the MFFC stands out among the recent advanced methods with an accuracy of 98.72% and an average cost of 5.34 milliseconds on the Malimg dataset. Our method can effectively identify malware and detect variants of malware families, which has excellent feature extraction capability and higher accuracy with lower detection time.

Malicious Code Variant Detection : A Survey

Malicious codes have become one of the major threats to computer systems. The malicious software which is also referred to as malware is designed by the attackers and can change their code as they propagate. The existing defense against malware is highly affected by the diversity and volume of malware variants that are being created rapidly. The variants of malware families exhibit typical behavioral patterns exhibiting their origin and purpose. The behavioral patterns can be exploited statically or dynamically to detect and classify malware into their known families. This paper provides a detailed survey of techniques to detect and classify malware into their respective families.

A'ãma Mrémé

A’ãma mrémé, or a’ãma speech, spoken by incumbents of a special ceremonial role within central Brazilian A’uwẽ-Xavante society, illustrates Joel Sherzer’s argument regarding the centrality of speech play to both linguistic and social analysis. A’ãma mrémé is a ludic code variant, a system of lexical substitution primarily at the level of nouns and verbs, spoken exclusively by a’ãma. Morphological analysis of of a’ãma mrémé reveals the existence of conceptual categories and perceptions that are not otherwise linguistically transparent. Further, a’ãma, who indexically make their ceremonial identity and role salient through the everyday practice of ãma speech, add an additional layer of complexity to A’uwẽ-Xavante’s complicated dualistic system of social organization. This complexity, heretofore overlooked by anthropologists of A’uwẽ-Xavante, becomes apparent through attention to socially situated discourse and verbal arts. A’ãma mrémé enriches and adds complexity to understandings of A’uwẽ-Xavante language, thought, and social organization. It is, as Sherzer contends, a place where language, cognition, perception, worldview and social structure come together in distilled form.

Building Pocket Code build-variants

Pocket Code is an integrated development environment (IDE) targeted at smartphones. With this IDE users can create mobile apps for the block-based visual programming language Catrobat. Pocket Code is released in various flavors with custom features for partners and projects (e.g., Pocket Code, Create@School, Phiro, and Standalone). All flavors extend a single common project codebase according to flavor specific requirements. The Standalone variants (debug and release) convert a Catrobat project into an Android application to install it independently and execute it without the need for an installed Pocket Code on an Android smartphone. Furthermore, it can be published on app stores for reputational and also monetary benefits. The app resource files and the configuration are generated on the fly upon a user request via the Pocket Code sharing platform. In this paper, the approach of building a Pocket Code variant and transform a Pocket Code project into an Android application are described. Especially the Standalone build variants have the potential to bring many interesting apps to the market.

How to build and use special purpose PC clusters in stellar dynamics

Large scale, direct particle-particle, brute force N-body simulations are required to accurately resolve numerically transport processes of energy and angular momentum due to two-body relaxation, and interactions between supermassive black holes and other particles having a much smaller mass. Direct accurate N-body codes are the widely used tool for such simulations, e.g., NBODY4 or NBODY6 (Aarseth 1999, 2003), see also Harfst et al. (2007) for a less complex code variant, used for benchmarks in this paper. Makino (2002) has presented another direct N-body summation code, which is optimized for a quadratic layout of processor (p required to be a square number).