Search

Security researchers utilize taint analyses to uncover suspicious behaviors in Android apps. Current static taint analyzers cannot handle ICC, reflection, and lifecycles dependably, increasing the result verification cost. On the other hand, current dynamic taint trackers accurately detect execution paths. However, they depend on specific Android versions and modified devices, reducing their usability and applicability. In addition, they require app exercise every time running the taint analysis. This paper presents a new dynamic taint tracker called T-Recs, tracking information flows by recording and reconstructing the app execution. First, before the taint analysis, the app’s runtime data are obtained by instrumenting logging code into the app’s bytecode and running the app to be independent of specific Android versions and devices. Then, T-Recs performs the taint analysis accurately with the logged data and separately from the app exercise. This paper is an extended version of our work published. Previously, T-Recs’ accuracy was mainly evaluated in privacy leak detection. The results show that T-Recs outperforms compared analyzers, which are FlowDroid (w/ and w/o IC3), Amandroid, DroidSafe, and TaintDroid (w/ and w/o IntelliDroid). This paper also involves DroidRA and IccTA. This paper shows that T-Recs detects ICC- and reflection-related leaks missed by FlowDroid in popular Google Play apps. The other static analyzers fail to analyze most of the apps. These experiments also indicate an advantage of T-Recs: its users can re-execute T-Recs’ taint analysis without re-exercising the app. T-Recs’ app-runtime overhead and parallel execution performance were also evaluated, and the results are acceptable.

Automatic machine translation plays an important role in reducing language barriers between people speaking different languages. Deep neural networks (DNN) have attained major success in diverse research fields such as computer vision, information retrieval, language modelling, and recently machine translation. Neural sequence-to-sequence networks have accomplished noteworthy progress for machine translation. Inspired by the success achieved by residual connections in different applications, in this work, we introduce a novel NMT model that adopts residual connections to achieve better performing automatic translation. Evaluation of the proposed model has shown an improvement in translation accuracy by 0.3 BLEU compared to the original model, using an ensemble of 5 LSTMs. Regarding training time complexity, the proposed model saves about 33% of the time needed by the original model to train datasets of short sentences. Deeper neural networks of the proposed model have shown a good performance in dealing with the vanishing/exploding problems. All experiments have been performed over NVIDIA Tesla V100 32G Passive GPU and using the WMT14 English-German translation task.

Panel data, which consist of observations on many individual units over two or more instances of time, have gradually become an important type of scientific data. Subsequently causal inference for panel data has attracted enormous interest from many fields as well as statistics. In this study, the rigorously formulated information flow analysis for time series, which is very concise in form and has been successfully applied in different disciplines, is generalized to identify the causality from homogeneous and independent identically distributed panel data. The resulting formula bears the same form as that for the former, though the meanings of the symbols differ. An algorithm is then proposed for panel data causality analysis, which has been validated with both linear and nonlinear problems. It has also been put to application to examine the causal relations among economic growth, energy consumption, trade openness, and energy price based on 15 Asian countries. Clearly identified are a strong bidirectional causality between economic growth and energy consumption, and a strong causality from import and export trade to economic growth. Energy price has no direct impact on energy consumption; it, instead, exerts a weak effect on the latter through influencing economic growth.

A cyber-physical system (CPS) represents the coupling network between a communication network and a power grid. Information flow (IF) as the medium of communication networks has an important influence on the performance assessment of this type of coupling network. However, traditional IF models cannot be applied to CPS directly because they do not take CPS distinctive coupling effect and associated operational mechanism into account. In this paper, we propose an information flow model (IFM) based on graph theory and traffic dynamics considering the functional architecture and transmission mode in realistic CPS to bridge the gap mentioned above. In the IFM, congestion rate and loss rate are employed to assess the performance of communication networks serving power grids under different transmission strengths and modes. Moreover, an auxiliary model about CPS topology is constructed, which can generate more realistic CPS topology. Case studies on an IEEE 39-bus system reveal the transmission performance of CPS in detail. The IFM is exploited to detect critical nodes and compare the effects of node enhancement on IF transmission of networks. In addition, IFM is also applied to compare the transmission performances of six network topologies, including three basic complex networks and three community-based networks. The results demonstrate the effectiveness of the proposed model, and some suggestions are provided on performance improvement of IF transmission for CPS.

After the military communications network is attacked by the enemy, in order to ensure that the information flow in the command and control organization can continue to be transmitted smoothly, the damaged network topology needs to be repaired in a timely and effective manner. The problem of repairing military communication network under the circumstance of nodes are hit in the network is described. This paper propose a method to repair damaged network by adding edges to the damaged network topology, which aims at maximizing network invulnerability. Network topological connection cost is based on average connection degree. The repair model is established with the consideration of connect cost limit and network connectivity. The model solving method based on discrete artificial bee colony algorithm is designed. Finally, proposed repair method is tested and compared with other three edge addition repair methods with a specific case of network. The experimental results show that the proposed method could enhance damaged network's invulnerability and outperforms the compared methods.

Naive Bayes classifier (NBC) is an effective classification technique in data mining and machine learning, which is based on the attribute conditional independence assumption. However, this assumption rarely holds true in real-world applications, so numerous researches have been made to alleviate the assumption by attribute weighting. To the best of our knowledge, almost all studies have calculated attribute weights according to correlation measure or classification accuracy. In this paper, we propose a novel causality-based attribute weighting method to establish the weighted NBC called IFG-WNBC, where causal information flow (IF) theory and genetic algorithm (GA) are adopted to search for optimal weights. The introduction of IF produces a bran-new weight measure criterion from the angle of causality other than correlation. The population initialization in GA is also improved with IF-based weights for efficient optimization. Multi-set of comparison experiments on UCI data sets demonstrate that IFG-WNBC achieves superiority over classic NBC and other common weighted NBC algorithms in classification accuracy and running time.

Using molecular profiles to predict the drug response is of practical importance in precision medicine and have been extensively studied. Due to the complexity of molecular information, the existing methods couldn't capture enough information and their prediction performances are not satisfying. In this study, we propose a method called DIMDRP (double iteration method for drug response prediction) which improves a lot in the prediction accuracy. DIMDRP integrates several important molecular information including miRNA expression, drug chemical structure, target interaction, drug-target interaction and cell line-drug response, and constructs a heterogeneous network. Then an improved information flow iteration algorithm is used to calculate association scores of cell line-drug responses, and we prioritize the cell lines for each query drug. The cross-validation experiments show that the average area under curve (AUC) of DIMDRP is as high as 0.8953, and two other measurement metrics are applied to assess the performance. When compared to other approaches, our method shows a significant advantage for all metrics. We also consider the specific tissue condition into our competition from a practical aspect. Last, a case is studied to predict novel cell line-drug responses, some of which can be evidenced by previous experiments.

Online social networking services (SNSs) have emerged rapidly and have become huge data sources for social network analysis. The spread of the content generated by users is crucial in SNS, but there is only a handful of research works on information diffusion and, more precisely, information diffusion flow. In this paper, we propose a novel method to discover information diffusion processes from SNS data. The method starts preprocessing the SNS data using a user-centric algorithm of community detection based on modularity maximization with the purpose of reducing the complexity of the noisy data. After that, the InfoFlow miner generates information diffusion flow models among the user communities discovered from the data. The algorithm is an extension of a traditional process discovery technique called the Flexible Heuristics miner, but the visualization ability of the generated process model is improved with a new measure called response weight, which effectively captures and represents the interactions among communities. An experiment with Facebook data was conducted, and information flow among user communities was visualized. Additionally, a quality assessment of the models was carried out to demonstrate the effectiveness of the method. The final constructed models allowed us to identify useful information such as how the information flows between communities and information disseminators and receptors within communities.

Effort has been made to find biomarkers for vascular dementia (VaD). Nevertheless, the current findings are typically obtained through statistical tests of group level differences. In clinical practice, however, it is more common to perform individual level inferences, e.g., to determine if a subject is suffering from VaD, which cannot be resolved with statistical analysis. The goal of this study is to develop a method to effectively discriminate early VaD patients from normal controls by combining EEG features with machine learning methods. The EEG signals were recorded from a total of 15 VaD patients and 21 controls during a visual oddball task. Interregional directed connectivity was derived from directed transfer function (DTF) analysis and used as features in classification. Three machine learning methods, including linear discriminant analysis (LDA), error back-propagation (BP) neural network, and support vector machine (SVM) were used as classifiers, and their classification performance was compared. It was found that VaD patients can be effectively identified using the BP and SVM classifiers with high accuracy. In particular, when the SVM classifier was combined with feature selection by Fisher score, it reached an accuracy 86.11%, sensitivity 86.67%, and specificity 85.71%. The area under the curve (AUC, 0.854) indicates a good identification of VaD patients from the normal controls. Since the EEG is noninvasive, inexpensive, and widely available to use, the current study presents a novel clinical application of machine learning methods and could facilitate automatic screening and diagnosis of the VaD at an early stage in future.

This paper proposes a novel image feature set based on a principled information theoretic analysis of the convolutional neural network (CNN). The output of convolutional filters is modeled as a random variable conditioned on the object class and network filter bank. The conditional entropy (CENT) of filter outputs is shown in theory and experiments to be a highly compact and class-informative feature that can be computed from the CNN feature maps and used to obtain higher classification accuracy than the original CNN itself. Experiments involve three binary classification tasks using the 3D brain MRI data: Alzheimer's disease (AD) versus healthy controls (HC), young versus old age, and male versus female, where the area under the curve (AUC) values for the CENT feature classification (93.9%, 96.7%, and 71.9%) are significantly higher than the softmax output of the original CNN classifier trained for the task (81.6%, 79.4%, and 63.1%). A statistical analysis based on the Wilcoxon test identifies CENT features with significant links to brain labels, which could potentially serve as diagnostic biomarkers.

Background: In order to make informed decisions, software engineering practitioners need information from testing. However, with the trend of increased automation, there is exponential growth and increased distribution of this information. This paper aims at exploring the information flow in software testing in the domain of embedded systems. Method: Data was collected through semi-structured interviews of twelve experienced practitioners with an average work experience of more than fourteen years working at five organizations in the embedded software industry in Sweden. Seventeen hours of audio recordings were transcribed and anonymized into 130 pages of text that was analyzed by means of thematic analysis. Results: The flow of information in software testing can be represented as feedback loops involving stakeholders, software artifacts, test equipment, and test results. The six themes that affect the information flow are how organizations conduct testing and trouble shooting, communication, processes, technology, artifacts, and the organization of the company. Seven main challenges for the flow of information in software testing are comprehending the objectives and details of testing; root cause identification; poor feedback; postponed testing; poor artifacts and traceability; poor tools and test infrastructure; and distances. Finally, five proposed approaches for enhancing the flow are: close collaboration between roles; fast feedback; custom test report automation; test results visualization; and the use of suitable tools and frameworks. Conclusions: The results indicate that there are many opportunities to improve the flow of information in software testing: a first mitigation step is to better understand the challenges and approaches. Future work is needed to realize this in practice, for example, on how to shorten feedback cycles between roles, as well as how to enhance exploration and visualization of test results.

Vast quantities of data are generated by social networks in seconds. The information generated in a social network is transformed into a flow by the subjects who produce, transmit, and consume it. This flow can be represented in a very complicated directional graph where each subject is represented as a node, and the flow of information is represented as a directed edge. In this paper, we introduce a method of dividing this complex directional graph by user and quantifying the flow of information between and among users based on information flow vectors. We propose a system that can monitor the flow of information in social networks using information flow vectors extracted from social media data. We also introduce an improved skyline algorithm that can respond quickly to a user's various queries.

Correlation analysis is of great significance for exploring the multivariate data sets as it helps researchers toward an in-depth understanding of the complex interactions and relationships among variables. In this paper, we propose a correlation analysis method that identifies salient scalars for multivariate data exploration. We exploit specific mutual information metric to measure the information overlap and analyze the relationships between one scalar and other variables. Moreover, we define the information flow and introduce another metric, influence to quantify the associations among scalars of different variables. Furthermore, we integrate these two information metrics and construct a surprise-influence map for users' interaction to identify the salient scalars. By investigating the relationships among these salient scalars, we analyze the correlations among variables. We demonstrate the applicability and effectiveness of our proposed method by applying it to different data sets.

This paper aims to explore the automatic detection method of epileptic seizures to improve the treatment and diagnosis of medically refractory epilepsy patients. Anew algorithm based on directed transfer function (DTF) method was proposed for epileptic seizure detection. First, the sliding window technique was used to segment electroencephalogram (EEG) recordings, and the cerebral functional connectivity was calculated by the DTF algorithm. Then, the total information outflow based on the DTF-derived connectivity was calculated by adding up the information flow from a single EEG channel to other channels. Finally, the information outflow was assigned as the features of support vector machine (SVM) classifier to discriminate interictal and ictal EEG segments. For 10 epilepsy patients, the proposed algorithm provided the mean correct rate of 98.45%, the mean selectivity of 64.43%, the mean sensitivity of 93.36%, the mean specificity of 98.42%, and the average detection rate of 95.89%. By applying the statistical analysis, the superiority of DTF-based method was statistically significant when compared with other algorithms in terms of five assessment criteria. Our results indicated that the DTF-derived connectivity could characterize the dynamic causal interaction patterns between brain areas during seizure states, and the proposed method was suitable for the detection of epileptic seizures.

Trusted hardware sharing (THS) system can provide multiple trusted execution environments (TEE) via sharing the trusted hardware (e.g., sharing trusted platform module via virtualization) for stand-alone and isolation scenarios. However, the trusted function requests (TFRs) sent to the trusted hardware are emitted by multiple TEEs, which have to be processed by THS. Since different applications in different TEEs have different security requirements, the data in TFRs need to be protected from being leaked or modified in an unauthorized manner. To address this issue, we present a secure scheme for THS systems based on an information flow model that protects the sensitive data in TFRs. Each TFR is assigned a security level according to their owner, and processed in isolated environments with different security levels. We implement the prototype and conduct the experiments in both shared memory and isolated environments. The results indicate that the introduction of security mechanisms can lead to more time consumption on processing TFRs with the increase in the dimension of security levels. However, this degradation in performance is still acceptable and can be mitigated in the real world, because intensive TFR requests are not present as they are in the experimental environment.