Search

Machine learning has been widely employed in the literature of malware detection because it is adapted to the need for scalability in vetting large scale samples of Android. Feature engineering has therefore been the key focus for research advances. Recently, a new research direction that builds on the momentum of Deep Learning for computer vision has produced promising results with image representations of Android byte- code. In this work, we postulate that other artifacts such as the binary (native) code and metadata/configuration files could be looked at to build more exhaustive representations of Android apps. We show that binary code and metadata files can also provide relevant information for Android malware detection, i.e., that they can allow to detect Malware that are not detected by models built only on bytecode. Furthermore, we investigate the potential benefits of combining all these artifacts into a unique representation with a strong signal for reasoning about maliciousness.

Background The availability of high-throughput omics datasets from large patient cohorts has allowed the development of methods that aim at predicting patient clinical outcomes, such as survival and disease recurrence. Such methods are also important to better understand the biological mechanisms underlying disease etiology and development, as well as treatment responses. Recently, different predictive models, relying on distinct algorithms (including Support Vector Machines and Random Forests) have been investigated. In this context, deep learning strategies are of special interest due to their demonstrated superior performance over a wide range of problems and datasets. One of the main challenges of such strategies is the “small n large p” problem. Indeed, omics datasets typically consist of small numbers of samples and large numbers of features relative to typical deep learning datasets. Neural networks usually tackle this problem through feature selection or by including additional constraints during the learning process. Methods We propose to tackle this problem with a novel strategy that relies on a graph-based method for feature extraction, coupled with a deep neural network for clinical outcome prediction. The omics data are first represented as graphs whose nodes represent patients, and edges represent correlations between the patients’ omics profiles. Topological features, such as centralities, are then extracted from these graphs for every node. Lastly, these features are used as input to train and test various classifiers. Results We apply this strategy to four neuroblastoma datasets and observe that models based on neural networks are more accurate than state of the art models (DNN: 85%-87%, SVM/RF: 75%-82%). We explore how different parameters and configurations are selected in order to overcome the effects of the small data problem as well as the curse of dimensionality. Conclusions Our results indicate that the deep neural networks capture complex fe

We demonstrate the potential for the Global Positioning System (GPS) to provide highly accurate tropospheric products for use in meteorological applications. Tropospheric products, in particular the wet delays, are treated as an unknown parameter in GPS processing and are estimated with other parameters such as station coordinates. In this study, we investigate the effects of Phase Center Correction (PCC) models on tropospheric zenith wet delay (ZWD), integrated water vapor (IWV) and gradient products. Two solutions were generated using the GAMIT software over the EUREF Permanent GNSS Network (EPN). The first (reference) solution was derived by applying the International GNSS Service (IGS) type-mean PCC model, while for the second solution PCC models from individual calibrations were used. The solutions were generated identically, except for the PCC model differences. The two solutions were compared, with the assumption that common signals are differenced out. The comparison of the two solutions clearly shows a bias in all tropospheric products, which can be attributed to PCC model deficiencies. Overall, mean biases of ±1.8, ±0.3, ±0.14 and ±0.19 mm are evident in ZWD, IWV, North-South and East-West gradients, respectively. Moreover, the differences between the two solutions show seasonal variations. For all antenna types, the ZWD and IWV differences are dominated by white plus power-law noise, with latter characterizing the low-frequency spectrum. On the other hand, the horizontal gradients exhibit a white plus first order autoregressive noise characteristic with less than 1% white noise. The individual PCC model provides a better fit to an external independent model in terms of gradient estimate and also provides up to 3 % more carrier phase ambiguity resolution.

The thesis analyses the judicial review of laws, more precisely their compatibility with "higher law" in the Grand Duchy of Luxembourg. Since the 19th century, general courts have authority to review laws as to international law. In 1996/1997, the Constitutional Court was established and empowered to examine ordinary statute law as to their compatibility with the Constitution. As a result, the rule of law might be considered realized. However, the powers of this special court and the legal effects of its judgements are considerably limited. What is more, the analysis of constitutional jurisprudence reveals clear self-limiting tendancies of the constitutional judges themselves. Moreover, it is general wisdom that the Constitutional Court has no authority to review the compatibility of any law with international law. The latter competence is exclusively with the ordinary judge. This situation causes legal uncertainty, bearing the risk of contradictory jurisprudence of high courts. The judicial separation of constitutionality and conventionality review also lacks coherence with respect to the primacy of public international in the Grand-Duchy, and it ignores the existing overlapping of international law and Constitution., La présente étude porte sur le contrôle juridictionnel des lois, plus précisément sur la compatibilité des lois avec le "droit supérieur" du Grand-Duché de Luxembourg. Depuis le XIXe siècle, le juge ordinaire est compétent pour vérifier la compatiblité des lois avec le droit international. En 1996/1997, la Cour constitutionnelle a été instaurée et chargée de contrôler la constitutionnalité des lois. En conséquence, l'on pourrait être amené à considérer l'État de droit comme réalisé au Luxembourg. Cependant, les compétences et les effets juridiques des décisions de la Cour constitutionnelle sont considérablement limités. Qui plus est, l'analyse de la jurisprudence constitutionnelle permet de reveler des tendances d'autolimitation des juges constitutionnels. De surcroît, il est généralement reconnu que la Cour constitutionnelle ne saurait examiner la conventionnalité des lois. Le juge ordinaire est exclusivement compétent pour procéder à ce contrôle. Cette situation est source d'insécurité juridique, en ce qu'elle risque d'engendrer une jurisprudence divergente entre les hautes cours. Par ailleurs, la séparation du contrôle de conventionnalité de celui de constitutionnalité n'est guère cohérent au regard de la reconnaissance de la règle de primauté du droit international sur le droit interne dans son intégralité au Grand-Duché. Cette séparation ne tient pas compte de l'imbrication des normes issues du droit international et de la Constitution.

The evaluation of extraction protocols for untargeted metabolomics approaches is still difficult. We have applied a novel stable isotope-assisted workflow for untargeted LC-HRMS-based plant metabolomics , which allows for the first time every detected feature to be considered for method evaluation. The efficiency and complementarity of commonly used extraction solvents, namely 1 + 3 (v/v) mixtures of water and selected organic solvents (methanol, acetonitrile or methanol/acetonitrile 1 + 1 (v/v)), with and without the addition of 0.1% (v/v) formic acid were compared. Four different wheat organs were sampled, extracted and analysed by LC-HRMS. Data evaluation was performed with the in-house-developed MetExtract II software and R. With all tested solvents a total of 871 metabolites were extracted in ear, 785 in stem, 733 in leaf and 517 in root samples, respectively. Between 48% (stem) and 57% (ear) of the metabolites detected in a particular organ were found with all extraction mixtures, and 127 of 996 metabolites were consistently shared between all extraction agent/organ combinations. In aqueous methanol, acidification with formic acid led to pronounced pH dependency regarding the precision of metabolite abundance and the number of detectable metabolites, whereas extracts of acetonitrile-containing mixtures were less affected. Moreover, methanol and acetonitrile have been found to be complementary with respect to extraction efficiency. Interestingly, the beneficial properties of both solvents can be combined by the use of a water-methanol-acetonitrile mixture for global metabolite extraction instead of aqueous methanol or aqueous acetonitrile alone.

The continuous evolution of global navigation satellite systems (GNSS) meteorology has led to an increased use of associated observations for operational modern low-latency numerical weather prediction (NWP) models, which assimilate GNSS-derived zenith total delay (ZTD) estimates. The development of NWP models with faster assimilation cycles, e.g., 1-h assimilation cycle in the rapid update cycle NWP model, has increased the interest of the meteorological community toward sub-hour ZTD estimates. The suitability of real-time ZTD estimates obtained from three different precise point positioning software packages has been assessed by comparing them with the state-of-the-art IGS final troposphere product as well as collocated radiosonde (RS) observations. The ZTD estimates obtained by BNC2.7 show a mean bias of 0.21 cm, and those obtained by the G-Nut/Tefnut software library show a mean bias of 1.09 cm to the IGS final troposphere product. In comparison with the RS-based ZTD, the BNC2.7 solutions show mean biases between 1 and 2 cm, whereas the G-Nut/Tefnut solutions show mean biases between 2 and 3 cm with the RS-based ZTD, and the ambiguity float and ambiguity fixed solutions obtained by PPPWizard have mean biases between 6 and 7 cm with the references. The large biases in the time series from PPP-Wizard are due to the fact that this software has been developed for kinematic applications and hence does not apply receiver antenna eccentricity and phase center offset (PCO) corrections on the observations. Application of the eccentricity and PCO corrections to the a priori coordinates has resulted in a 66 % reduction of bias in the PPP-Wizard solutions. The biases are found to be stable over the whole period of the comparison, which are criteria (rather than the magnitude of the bias) for the suitability of ZTD estimates for use in NWP nowcasting. A millimeter-level impact on the ZTD estimates has also been observed in relation to ambiguity resolution. As a result of a c

Le monarque et les membres de la famille régnante ont-ils le droit de vote, actif et passif ? Font-ils partie du « peuple » ? Ce livre est la première monographie de science juridique en Europe à creuser ce sujet, d'habitude délaissé. Sous cette question anodine se cache, pourtant, une clé essentielle pour cerner la nature de la monarchie constitutionnelle et son processus de démocratisation. Le monarque est-il toujours un « être à part », séparé du peuple - au Japon, d'antan, le tennô était considéré comme un dieu vivant -, ou est-il un« citoyen », ainsi que l'a revendiqué en 2004 Henri de Nassau, grand-duc de Luxembourg ? Du cas luxembourgeois, qui, sur ce sujet, est particulièrement intéressant, ce livre propose une étude serrée grâce à l'exhumation des listes électorales (1841-2013) et à l'élucidation de leurs fondements juridiques. Est mise à jour ainsi une rupture opérée en 1945 dont il convient de souligner à la fois l'aspect moderne (voire précurseur) et incomplet (et partant inconstitutionnel). Au-delà du cas luxembourgeois, le livre restitue la problématique dans son contexte général, théorique, historique et de droit comparé. De cette vaste fresque, portant sur plus de deux siècles et une dizaine de pays, ressort de nos jours surtout un modèle : celui du « citoyen monarque ». Ce modèle éclaire une nouvelle réalité de la monarchie : non pas celle, en déclin, des pouvoirs du chef d Etat, mais celle, en pleine ascension, de la personnalité de l'individu qui se trouve « derrière » l'organe, et qui, par un retournement historique spectaculaire, réclame d'être d'abord un être humain doté des droits de l'homme et un citoyen doté des droits politiques.

To face continuously growing security threats and requirements, sound methodologies for constructing secure systems are required. In this context, Model-Driven Security (MDS) has emerged since more than a decade ago as a specialized Model-Driven Engineering approach for supporting the development of secure systems. MDS aims at improving the productivity of the development process and quality of the resulting secure systems, with models as the main artifact. This paper presents how we systematically examined existing published work in MDS and its results. The systematic review process, which is based on a formally designed review protocol, allowed us to identify, classify, and evaluate different MDS approaches. To be more specific, from thousands of relevant papers found, a final set of the most relevant MDS publications has been identified, strictly selected, and reviewed. We present a taxonomy for MDS, which is used to synthesize data in order to classify and evaluate the selected MDS approaches. The results draw a wide picture of existing MDS research showing the current status of the key aspects in MDS as well as the identified most relevant MDS approaches.We discuss the main limitations of the existing MDS approaches and suggest some potential research directions based on these insights.