A graph-based model for malware detection and classification using system-call groups

In this paper we present a graph-based model that, utilizing relations between groups of System-calls, detects whether an unknown software sample is malicious or benign, and classifies a malicious software to one of a set of known malware families. More precisely, we utilize the System-call Dependency Graphs (or, for short, ScD-graphs), obtained by traces captured through dynamic taint analysis. We design our model to be resistant against strong mutations applying our detection and classification techniques on a weighted directed graph, namely Group Relation Graph, or Gr-graph for short, resulting from ScD-graph after grouping disjoint subsets of its vertices. For the detection process, we propose the \(\Delta \)-similarity metric, and for the process of classification, we propose the SaMe-similarity and NP-similarity metrics consisting the SaMe-NP similarity. Finally, we evaluate our model for malware detection and classification showing its potentials against malicious software measuring its detection rates and classification accuracy.     

ETHYLENE OXIDATION ON SILVER CATALYSTS: EFFECT OF ETHYLENE OXIDE AND OF EXTERNAL TRANSFER LIMITATIONS

Ethylene oxidation to ethylene oxide and to carbon dioxide over silver catalysts was studied in a CSTR. The effects of two factors on the catalyst performance were examined. The first was the presence of excess ethylene oxide in the feed. A kinetic model was introduced which assumed that ethylene and ethylene oxide compete for the same sites on the catalyst surface. This model provided reasonable quantitative agreement with kinetic and potentiometric measurements. The second factor that was studied was the presence of external heat and mass transfer limitations. It was found that such limitations cause a significant decrease of the selectivity to ethylene oxide. This decrease is a result of the temperature difference between the catalyst surface and the bulk of the gas phase and of the fact that the activation energy of ethylene combustion is greater than that of ethylene epoxidation. The contribution of other factors such as inhibition by CO₂ or possible incomplete mixing in the reactor is shown to be insignificant.     

Performance vs. freshness in web database applications

In this work, we present a novel approach for the efficient materialization of dynamic web pages in e-commerce applications such as an online retail store with millions of items, hundreds of HTTP requests per second and tens of dynamic web page types. In such applications, user satisfaction, as measured in terms of response time (QoS) and content freshness (QoD), determines their success especially under heavy workload. The novelty of our materialization approach over existing ones is that, it considers the data dependencies between content fragments of a dynamic web page. We introduce two new semantic-based data freshness metrics that capture the content dependencies and propose two materialization algorithms that balance QoS and QoD. In our evaluation, we use a real-world experimental system that resembles an online bookstore and show that our approach outperforms existing QoS-QoD balancing approaches in terms of server-side response time (throughput), data freshness and scalability.     

Extraction and identification of natural antioxidant from the seeds of the Moringa oleifera tree variety of Malawi

The oil from the dried seeds of the Moringa oleifera tree (variety of Malawi) was extracted with a mixture of chloroform/methanol (50∶50). The induction period measurements demonstrated a great resistance to oxidative rancidity. After degumming, there was a reduction of 74% in induction periods. The gums produced were extracted with diethylether, n-butanol, and water, yielding four fractions: Fraction 1 (81.8% w/w), Fraction 2 (0.04% w/w), Fraction 3 (0.05% w/w), and Fraction 4 (17.0% w/w). These fractions were tested for their protection of fresh sunflower oil against rancidity, at 50°C, using a UV accelerated method. The oxidation of the sunflower oil was measured using PV; absorbance E _1cm ^1% and malondialdehyde concentration were measured by HPLC. The fraction that showed the highest antioxidant activity was further fractionated by HPLC, yielding seven fractions. Fraction HPLC 3 (present in a quantity of 330.8 and 29.11 ppm in gums and oil, respectively) showed the highest antioxidant activity. Its activity was also compared with that of the commonly used antioxidants BHT and α-tocopherol on sunflower oil using the same methods. At the same level of addition (200 ppm), HPLC 3 showed higher antioxidant activity than BHT and α-tocopherol. The identification of HPLC3 was done using ¹H NMR, ¹³C NMR, MS, melting point, and UV absorption spectroscopy and proved to be 3,5,7,3′,4′,5′-hexahydroxyflavone (myricetin).     

The Longest Path Problem has a Polynomial Solution on Interval Graphs

The longest path problem is the problem of finding a path of maximum length in a graph. Polynomial solutions for this problem are known only for small classes of graphs, while it is NP-hard on general graphs, as it is a generalization of the Hamiltonian path problem. Motivated by the work of Uehara and Uno (Proc. of the 15th Annual International Symp. on Algorithms and Computation (ISAAC), LNCS, vol. 3341, pp. 871–883, 2004), where they left the longest path problem open for the class of interval graphs, in this paper we show that the problem can be solved in polynomial time on interval graphs. The proposed algorithm uses a dynamic programming approach and runs in O(n ⁴) time, where n is the number of vertices of the input graph.     

Multimodal and ontology-based fusion approaches of audio and visual processing for violence detection in movies

In this paper we present our research results towards the detection of violent scenes in movies, employing advanced fusion methodologies, based on learning, knowledge representation and reasoning. Towards this goal, a multi-step approach is followed: initially, automated audio and visual analysis is performed to extract audio and visual cues. Then, two different fusion approaches are deployed: (i) a multimodal one that provides binary decisions on the existence of violence or not, employing machine learning techniques, (ii) an ontological and reasoning one, that combines the audio-visual cues with violence and multimedia ontologies. The latter reasons out not only the existence of violence or not in a video scene, but also the type of violence (fight, screams, gunshots). Both approaches are experimentally tested, validated and compared for the binary decision problem of violence detection. Finally, results for the violence type identification are presented for the ontological fusion approach. For evaluation purposes, a large dataset of real movie data has been populated.     

On the use of the BLYP functional for the DFT calculation of graphite–hydrogen systems

This letter focuses on a possible pitfall that can occur in the DFT calculation of graphite–hydrogen interactions under tokamaks conditions. Calculations based on the BLYP functional, in fact, result in a stable –C₂H₃ group that other functionals (OLYP, XLYP, PBE and BP) do not confirm.     

Analytical Approaches to Stress Intensity Factor Evaluation for Indentation Cracks

In the present paper, an analytical solution for the stress intensity factor in the case of cracks produced by Vickers indenters has been extended to the cases of triangular indenters, i.e., Berkovich and cube-corner. According to the adopted approach, median/radial cracks produced by indentations are modeled as loaded by either a point-force or a symmetric disk-shaped wedge. The wedge diameter is assumed to be equal to the plastic zone size whereas the wedge thickness is evaluated by comparing the wedge volume with the hardness-impression volume. The point-force and the disk-shaped wedge analyses produce an upper and a lower bound, respectively, of the geometry-dependent parameter appearing in the expression for the fracture toughness. The predictions of the present analysis are in good agreement with similar experimental and numerical results.     

Development of a fast,lean and agile direct pelletization process using experimental design techniques

A novel hot melt direct pelletization method was developed, characterized and optimized, using statistical thinking and experimental design tools. Mixtures of carnauba wax (CW) and HPMC K100M were spheronized using melted gelucire 50–13 as a binding material (BM). Experimentation was performed sequentially; a fractional factorial design was set up initially to screen the factors affecting the process, namely spray rate, quantity of BM, rotor speed, type of rotor disk, lubricant–glidant presence, additional spheronization time, powder feeding rate and quantity. From the eight factors assessed, three were further studied during process optimization (spray rate, quantity of BM and powder feeding rate), at different ratios of the solid mixture of CW and HPMC K100M. The study demonstrated that the novel hot melt process is fast, efficient, reproducible and predictable. Therefore, it can be adopted in a lean and agile manufacturing setting for the production of flexible pellet dosage forms with various release rates easily customized between immediate and modified delivery.