Search

The end-to-end model has a wide range of applications in the fields of image recognition, natural language processing and speech recognition. The main benefit of this model is that the structure is a black box, and industrial users need only amend the inputs and outputs to obtain better performance for various applications. Nevertheless, in some contexts (e.g., medical-related tasks), the results of model generation require interpretability; furthermore, it is difficult to obtain the rich relationships of the hierarchical structure of the input, for the kind of data that are common in real life. In order to deal with these issues, this paper proposes an ontology-based deep belief network model to extend neural networks, using an approach that can be applied to hierarchical data. Multi-parameter data with hierarchical semantics are used in an experiment, and the results show that this model has the ability to handle hierarchical data and provide a certain level of interpretability.

Few-shot learning (FSL) is a core topic in the domain of machine learning (ML), in which the focus is on the use of small datasets to train the model. In recent years, there have been many important data-driven ML applications for intrusion detection. Despite these great achievements, however, gathering a large amount of reliable data remains expensive and time-consuming, or even impossible. In this regard, FSL has been shown to have advantages in terms of processing small, abnormal data samples in the huge application space of intrusion detection. FSL can improve ML for scarce data at three levels: the data, the model, and the algorithm levels. Previous knowledge plays an important role in all three approaches. Many promising methods such as data enrichment, the graph neural network model, and multitask learning have also been developed. In this paper, we present a comprehensive review of the latest research progress in the area of FSL. We first introduce the theoretical background to ML and FSL and then describe the general features, advantages, and main methods of FSL. FSL methods such as embedded learning, multitask learning, and generative models are applied to intrusion detection to improve the detection accuracy effectively. Then, the application of FSL to intrusion detection is reviewed in detail, including enriching the dataset by extracting intermediate features, using graph embedding and meta-learning methods to improve the model. Finally, the difficulties of this approach and its prospects for development in the field of intrusion detection are identified based on the previous discussion.

Iron carbide clusters are studied extensively in many catalysis fields by quantum chemistry methods. The high cost of structural energy calculation has always been the bottleneck of its development. First principles based on density functional theory (DFT) have high calculation accuracy and sound portability. Still, its computational cost is so high that it is difficult to carry out large-scale, long-term, and high-throughput simulations. A data driven potential is crucial for further accelerating the screening or prediction process in the high through-put calculation. In this paper, we generate 177k clusters data and choose some state-of-the-art machine learning models in physical chemistry to train them. The generated potential gives a very high prediction accuracy on the order of the structure stability and achieves better adaptability/tolerance on poor structures of clusters. In addition, we use the machine learning potential to assist in high-throughput data collection and realize the prediction of the adsorption position of hydrogen atoms on the cluster surface. We can get a more stable adsorption position of the hydrogen atom in a shorter time, compared with traditional quantum chemical calculators.

Iron sulfide nanoparticles play an essential role in the global biogeochemistry sulfur cycle and the origin of life. Controlling phase and morphology is essential but a great challenge for nanomaterials. Herein, we have investigated surface energy, morphology, thermodynamic stability, and phase evolution of FexSy nanoparticles through density functional theory calculations and Wulff construction. The sulfur-rich surfaces become more stable under higher sulfur chemical potential with low temperature and high partial H2S pressure, while the trends of sulfur-poor surfaces are reversed. Thermodynamic equilibrium phase diagrams imply that the intrinsic iron is sulfurated to mackinawite (FeS) and further to the polymorphs of FeS2 when the particle size is larger than 5 nm. When the size is decreased to 2 nm, the mackinawite (FeS) could be sulfurated to pyrrhotite (Fe7S8) and greigite (Fe3S4) and finally to the pyrite (FeS2) phase.

Information used in existing ontology matching solutions are usually grouped into four categories: lexical information, structural information, semantic information, and external information, respectively. By summarizing and analyzing the approaches for utilizing the same kind of information, this paper finds that lexical information is mainly analyzed based on text and dictionary similarity. Similarly, structural information and semantic information are mainly analyzed based on graph structure and reasoner, respectively. The approaches for aggregating information analysis results are discussed. Challenges in the analysis of various types of information for existing ontology matching solutions are also described, and insights into directions for future research are provided.

Fuzz testing has been successful in finding defects of various software packages. These defects include file parsing, image processing, Internet browsers, and network protocols. However, the quality of the initial seed test cases greatly influences the coverage and defect detection capability of fuzz testing. To address this issue, we propose CSEFuzz, a fuzz testing approach based on symbolic execution for defect detection. First, CSEFuzz generates candidate test cases by symbolic execution and collects coverage information of the test cases. Then, CSEFuzz extracts the test-case templates of the test cases and selects a set of test-case templates according to specific coverage criteria. Finally, CSEFuzz selects test cases according to the selected test-case templates, and the selected test cases are used as initial seed test cases for fuzz testing. Experiments are conducted on 11 open-source programs. The results show that in comparison with afl-cmin, which is the test-case selection command of Kelinci, CSEFuzz with a path coverage criterion reduces the time costs of the initial seed test selection and verification by 94.26%. In addition, compared with afl-cmin, 32 more paths are covered and 16 more defects are detected by CSEFuzz.

CHn is the precursor unit for graphene synthesis. We have theoretically predicated a series of CHn structures with n = 1, 2, 4, 6, 8, 10, and 12 at elevated pressures (ambient pressure, 50, 100, 200, 300, 350, and 400 GPa) using evolutionary algorithms. The predicted CH and CH2 structures are graphane-type and polyethylene over the whole considered pressure range, respectively. The molecular crystalline methane is predicted for the stoichiometry of CH4. The combination of methane and H2 for CH6, CH8, CH10, and CH12 up to 300 GPa are obtained. At 400 GPa, the mixture of polymer and H2 for CH6, CH10, and CH12 comes into play. From the computed enthalpy, higher pressure and more hydrogen concentration contributed to the decomposition (to carbon and H2) of CHn systems. The total density of states for these CHn structures show that only the CH12 phase is metallic above 300 GPa. The rotational properties are traced in H2 and the CHn structures. The CH4 rotation is more sensitive to the pressure. The H2 units are nearly freely rotational. Other structures of CHn, including fcc-type and experimentally known structures, are not competitive with the structures predicted by evolutionary algorithms under high pressure region. Our results suggest that the CHn (n > 4) system is a potential candidate for hydrogen storage where H2 could be released by controlling the pressure.

In 5G systems and beyond, traditional generic service models are no longer appropriate for highly customized and intelligent services. The process of reinventing service models involves allocating available resources, where the performance of service processes is determined by the activity node with the lowest service rate. This paper proposes a new bottleneck-aware resource allocation approach by formulating the resource allocation as a max-min problem. The approach can allocate resources proportional to the workload of each activity, which can guarantee that the service rates of activities within a process are equal or close-to-equal. Based on the business process simulator (i.e., BIMP) simulation results show that the approach is able to reduce the average cycle time and improve resource utilization, as compared to existing alternatives. The results also show that the approach can effectively mitigate the impact of bottleneck activity on the performance of service processes.

Deep neural networks have been used for traffic classifications and promising results have been obtained. However, most of the previous work confined to one specific task of the classification, where restricts the classifier potential performance and application areas. The traffic flow can be labeled from a different perspective which might help to improve the accuracy of classifier by exploring more meaningful latent features. In addition, deep neural network (DNN)-based model is hard to adapt the changes in new classification demand, because of training such a new model costing not only many computing resources but also lots of labeled data. For this purpose, we proposed a multi-output DNN model simultaneously learning multi-task traffic classifications. In this model, the common knowledge of traffic is exploited by the synergy among the tasks and improves the performance of each task separately. Also, it is showed that this structure shares the potential of meeting new demands in the future and meanwhile being able to achieve the classification with advanced speed and fair accuracy. One-shot learning, which refers to the learning process with scarce data, is also explored and our approach shows notable performance.

The mobile social network contains a large amount of information in a form of commentary. Effective analysis of the sentiment in the comments would help improve the recommendations in the mobile network. With the development of well-performing pretrained language models, the performance of sentiment classification task based on deep learning has seen new breakthroughs in the past decade. However, deep learning models suffer from poor interpretability, making it difficult to integrate sentiment knowledge into the model. This paper proposes a sentiment classification model based on the cascade of the BERT model and the adaptive sentiment dictionary. First, the pretrained BERT model is used to fine-tune with the training corpus, and the probability of sentiment classification in different categories is obtained through the softmax layer. Next, to allow a more effective comparison between the probabilities for the two classes, a nonlinearity is introduced in a form of positive-negative probability ratio, using the rule method based on sentiment dictionary to deal with the probability ratio below the threshold. This method of cascading the pretrained model and the semantic rules of the sentiment dictionary allows to utilize the advantages of both models. Different sized Chnsenticorp data sets are used to train the proposed model. Experimental results show that the Dict-BERT model is better than the BERT-only model, especially when the training set is relatively small. The improvement is obvious with the accuracy increase of 0.8%.

Considering the supply interruption in supply chain network, a closed-loop supply chain network model with dual objectives including total profit and total carbon emissions based on recycling mechanism was proposed. In this paper, we considered not only the elasticity of supply chain network but also the recycling prices. The model investigated the impact of the recycling mechanism on the supply interruption. The numerical results revealed the relationship between the degree of supply disruption and the recovery price, as well as the influence of the loss of suppliers’ supply capacity and the tolerance coefficient of shortage on the planning results.

Traditionally, chemistry problems are solved by means of a deductive approach. The question to be addressed is typically related to the value of a property that is either measured experimentally, computed using quantum-chemistry software, or (more recently) predicted using a machine-learned model. In this paper, we demonstrate that an inductive approach can be adopted using End-to-End (E2E) machine learning. This approach is illustrated for tackling the following chemistry problems: (i) determine the fully coordinated (FC) and undercoordinated (UC) atoms in a molecule with one missing atom, (ii) identify the type of atom that is missing in such an incomplete molecule, and (iii) predict the direction of a reaction between two molecules according to an existing dataset. The E2E approach leads to accuracies higher than 99%, 98%, and 93% for these three problems, respectively. Finally, in order to achieve such accuracies, a descriptor for the molecules, called bag of clusters, is introduced and compared with a series previously proposed descriptors, highlighting a series of advantages.

Distributed Ledger Technology (DLT) currently stands in the limelight. In many people's eyes, DLT has great potential to accelerate the digitalization process and substantially change the roles and services of central banks. On the other hands, as many new applications for DLT spring up, there isn't an assessment framework to evaluate DLT applied in the financial industry, leaving those systems to be easily under attack. This paper presents an initial DLT assessment framework for financial application. Through a suite of benchmarks, we show the evaluation results and explore the factors which affect the performance and security of DLT application.

Deep neural networks (DNNs) have made remarkable achievements in several areas, such as image classification. However, extensive researches show that DNNs are vulnerable to adversarial examples which are generated by adding some small perturbations on original images. DeepFool is an effective algorithm for generating adversarial examples with higher success rate and smaller perturbations. However, DeepFool computes the distance of all classification boundaries of the dataset, which causes the speed of DeepFool to be very slow. To address this problem, this paper proposes DADF, a dynamic adaptive method for generating adversarial examples. Based on tracing the information of the attacking process, DADF dynamically selects the subset of target labels. The subset of target labels is dynamically updated once successful adversarial examples are generated. Experimental results on MNIST and CIFAR10 datasets show that, comparing with DeepFool, DADF reduce the time greatly while generating 10000 adversarial examples on the MNIST and CIFAR10 dataset.

The application of machine learning programs is becoming increasingly more widespread. Neural networks, which are among the most popular machine learning programs, play important roles in people's daily lives, such as by controlling cars in autonomous driving systems. However, neural networks still lack effective testing methods. To address this problem, this paper proposes a testing method for neural network classifiers based on metamorphic relations. Firstly, it designs metamorphic relations to transform the original data set into derivative data sets. Then, it uses the data before and after the transformation to train and test the neural network classifier, respectively. Finally, it checks whether the output conforms to the metamorphic relations. The neural network classifier is defective if conflicts are detected. Experiments are conducted on a neural network classifier from Stanford's cs231n course to verify the effectiveness of the method. The results show that the defect detection capability of the proposed method is accurate, and 87.5% of the mutants are successfully detected.

At present, spectrum-based fault localization and mutation-based fault localization are the most popular fault localization methods. The two methods suffer from the problems that multiple statements are ranked the same and there are high time costs. To improve the performance of fault localization, this paper proposes a novel approach for fault localization based on integrating spectrum and mutation. In addition to decreasing the time consumption overhead of the mutation-based fault localization method, the approach improves the accuracy of the spectrum-based fault localization method. The experimental results show that the approach improves the accuracy of fault localization in comparison with the spectrum-based fault localization method.

Some Siamese-based trackers use temporal context prior as structural constraint to suppress background distractors. However, due to the lack of contour recognition, it is difficult to obtain a better performance. In order to address this issue, we propose a structural pixel-wise target attention strategy for robust object tracking with memory model. Firstly, a pixel-wise target attention model is constructed for evaluating the probability that the pixel belongs to the target, which can effectively discriminate the target boundary so as to highlight the target area. Meanwhile, structural information is used to solve pixel-wise distractors, which is combined with complementary pixel-wise label constraints to obtain a structural pixel-wise target attention model. This attention mechanism can improve the confidence of the final response map and achieve more reliable target location. Secondly, a memory model is learned using the highly reliable memory pattern for providing high-quality training samples for updating pixel-wise target attention model. Benefiting from this method, our method realizes the pixel-wise target attention model to adapt to the variation of the target while preventing the background noise, thus improving the discriminability of the model. Finally, the structural pixel-wise target attention mechanism and memory model are integrated into the Siamese-based tracking framework, which shows better merit for robust object tracking. Extensive experiments on multiple tracking benchmarks show that our approach achieves excellent performance in various challenging target tracking tasks.

Many ontologies are provided to representing semantic sensors data. However, heterogeneity exists in different sensors which makes some service operators of Internet of Thing (IoT) difficult (such as such as semantic inferring, non-linear inverted index establishing, service composing). There is a great deal of research about sensor ontology alignment dealing with the heterogeneity between the different sensor ontologies, but fewer solutions focus on exploiting syntaxes in a sensor ontology and the pattern of accessing alignments. Our solution infers alignments by extending structural subsumption algorithms to analyze syntaxes in a sensor ontology, and then combines the alignments with the SKOS model to construct the integration sensor ontology, which can be accessed via the IoT. The experiments show that the integration senor ontology in the SKOS model can be utilized via the IoT service, and the accuracy of our prototype, in average, is higher than others over the four real ontologies.

This paper focuses on matching ontologies created for similar domains through different sources. Different solutions use lexical, structural or logical processing and analysis to match ontologies. However, an important aspect is also interpreting concepts that entities are presented with and using them in relation to semantics in an ontology. The paper demonstrates analyzing and extending the concepts used to define entities in an ontology, discusses establishing and filtering matching candidates by reasoners, and then describes constructing correspondences between entities from different ontologies using lexical and semantic analysis. The experiments show that our prototype, called OMI-DL, is among the top group over many ontologies adopted from the OAEI benchmark data set. We also provide an evaluation of the OMI-DL method for matching the DOLCE+DnS Ultralite ontology and the domain ontology developed in the m:Ciudad project.

In this article, we design and evaluate the proposed Geographic-based Spray-and-Relay (GSaR) routing scheme in Delay/Disruption Tolerant Networks (DTNs). To the best of our knowledge, GSaR is the first spray based geographic routing scheme using the historical geographic information for making routing decision. Here, the term spray means only a limited number of message copies are allowed for replication in the network.\ud \ud By estimating a movement range of destination via the historical geographic information, GSaR expedites message being sprayed towards this range, meanwhile prevents that away from and postpones that out of this range. As such, the combination of them intends to fast and efficiently spray the limited number of message copies towards this range, and effectively spray them within range, in order to reduce the delivery delay and increase the delivery ratio. Furthermore, GSaR exploits Delegation Forwarding (DF) to enhance the reliability of routing decision and handle the local maximum problem, considered as the challenges for applying geographic routing scheme in sparse networks. We evaluate GSaR under three city scenarios abstracted from real world, with other routing schemes for comparison. Results show that GSaR is reliable for delivering messages before expiration deadline and efficient for achieving low routing overhead ratio. Further observation indicates that GSaR is also efficient in terms of a low and fair energy consumption over the nodes in the network.

Internet of Things is smartly changing various existing research areas into new themes, including smart health, smart home, smart industry, and smart transport. Relying on the basis of “smart transport,” Internet of Vehicles (IoV) is evolving as a new theme of research and development from vehicular ad hoc networks (VANETs). This paper presents a comprehensive framework of IoV with emphasis on layered architecture, protocol stack, network model, challenges, and future aspects. Specifically, following the background on the evolution of VANETs and motivation on IoV an overview of IoV is presented as the heterogeneous vehicular networks. The IoV includes five types of vehicular communications, namely, vehicle-to-vehicle, vehicle-to-roadside, vehicle-to-infrastructure of cellular networks, vehicle-to-personal devices, and vehicle-to-sensors. A five layered architecture of IoV is proposed considering functionalities and representations of each layer. A protocol stack for the layered architecture is structured considering management, operational, and security planes. A network model of IoV is proposed based on the three network elements, including cloud, connection, and client. The benefits of the design and development of IoV are highlighted by performing a qualitative comparison between IoV and VANETs. Finally, the challenges ahead for realizing IoV are discussed and future aspects of IoV are envisioned.

Efficient sensor deployment is one of the primary requirements of the precision agriculture use case of wireless sensor networks (WSNs) to provide qualitative and optimal coverage and connectivity. The application-based performance variations of the geometrical-model-based sensor deployment patterns restrict the generalization of a specific deployment pattern for all applications. Furthermore, single or double metrics-based evaluation of the deployment patterns focusing on theoretical or simulation aspects can be attributed to the difference in performance of real applications and the reported performance in the literature. In this context, this paper proposes a testbed-based multi-metric quality measurement of sensor deployment for the precision agriculture use case of WSNs. Specifically, seven metrics are derived for the qualitative measurement of sensor deployment patterns for precision agriculture. The seven metrics are quantified for four sensor deployment patterns to measure the quality of coverage and connectivity. Analytical- and simulation-based evaluations of the measurements are validated through testbed experiment-based evaluations which are carried out in “INDRIYA” WSNs testbed. Toward realistic research impact, the investigative evaluation of the geometrical-model-based deployment patterns presented in this paper could be useful for practitioners and researchers in developing performance guaranteed applications for precision agriculture and novel coverage and connectivity models for deployment patterns.