Search

In this paper, we build an autoregressive integrated moving average (ARIMA) model and analyze the short-term trends of the gold price (GP) and bitcoin price (BCP) based on historical data. In addition, we use MATLAB to perform statistics and analysis on the data, find that the probability of continuous depreciation or appreciation after day 5 decreases exponentially, and find the maximum and minimum fluctuations. Further, we build a trading strategy model that uses the Apriori algorithm to calculate the number of subsets where prices have risen or fallen for 5 consecutive days. Finally, we perform a sensitivity analysis on the established model.

Multivariate time series classification (MTSC) is a widely recognized and important task in machine learning. Current methods often fail to account for the correlation between holistic and local temporal features. In this paper, we propose a novel dual-scaled network called DSNet, which combines holistic and local features and merges temporal information through a dual-scaled gating mechanism. The DSNet model's holistic module introduces the encoder part of the native Transformer, while the local module mainly consists of our proposed residual convolution module (ResBlock) and Squeeze-And-Excite Block to extract local features. Our experimental results show that our proposed method achieves higher accuracy on 12 out of the 20 UEA MTSC archive datasets. Furthermore, our method has a higher average ranking of 1.55 compared to the other 7 state-of-the-art methods.

Daidzin, 4′, 7-dihydroxyisoflavone is an isoflavonic phytoestrogen present in leguminous plants. Traditional Chinese medicine utilizes daidzin to treat various diseases such diarrhea, fever, hepatitis, cardiac problems etc. In current study we examined the anticancer activity of daidzin against human cervical cancer in vitro. HeLa, human cervical cancer cell line was purchased from ATCC and the cells were cultured with DMEM medium. The cytotoxic effect of daidzin against HeLa cell line was analyzed with MTT assay. The IC-50 value was obtained at 20 µM hence the cells were treated with 20 µM of daidzin for further analysis. ROS generation was assessed with DCFH-DA staining and the induction of apoptosis was examined with Rhoadmine-123 staining. Acridine orange and ethidium bromide staining was done to examine the apoptotic and viable cells. Further the matrigel cell adhesion assay was done to analyze the inhibitory property of daidzin against cancer cell adhesion. Apoptotic induction of daidzin was examined by estimating the levels of Caspase 8 & 9 using ELISA technique. Inflammatory and cell proliferation signaling proteins were analyzed with qPCR analysis to confirm the anticancer activity of daidzin against human cervical cancer HeLa cell line. Daidzin significantly generated ROS and altered the mitochondrial membrane permeability in HeLa cell line. The results of AO/EtBr staining prove daidzin induced apoptosis in HeLa cell line and it also inhibited the cell adhesion property of HeLa which is reported in our matrigel cell adhesion assay. It also increased the caspases 8 & 9 which are key regulators of apoptosis. Daidzin significantly decreased the expression of inflammatory gene and cell proliferating signaling molecule. To, conclude our results confirm daidzin effectively decreased inflammation and induced apoptosis in human cervical cancer HeLa cell line.

Asphaltenes have a tendency toward aggregation, leading to a series of problems (e.g., coking) in the field of refining or other oil utilization and making the development of a disaggregation metho...

This paper proposes an Effiective biogeography-based optimization (EBBO) algorithm for solving the flow shop scheduling problem with intermediate buffiers to minimize the Total flow time (TFT). Discrete job permutations are used to represent individuals in the EBBO so the discrete problem can be solved directly. The NEH heuristic and NEH-WPT heuristic are used for population initialization to guarantee the diversity of the solution. Migration and mutation rates are improved to accelerate the search process. An improved migration operation using a two-points method and mutation operation using inverse rules are developed to prevent illegal solutions. A new local search algorithm is proposed for embedding into the EBBO algorithm to enhance local search capability. Computational simulations and comparisons demonstrated the superiority of the proposed EBBO algorithm in solving the flow shop scheduling problem with intermediate buffiers with the TFT criterion.

This paper proposes a character-level attention convolutional neural networks model (ACNN) for short-text classification task. The model is implemented on the deep learning framework which named tensorflow. The model can achieve better short-text classification result. Experimental datasets are from three different categories and scales. ACNN model are compared with traditional model such as LSTM and CNN. The experimental results show that ACNN model significantly improves the short-text classification results.

There is evidence that synaptic plasticity is a vital feature of realistic neuronal systems. This study, describing synaptic plasticity by a modified Oja learning rule, focuses on the effect of synapse learning rate on spike synchronization and its relative transitions in a Newman-Watts small-world neuronal network. The individual dynamics of each neuron is modeled by a simple Rulkov map that produces spiking behavior. Numerical results have indicated that large coupling can lead to a spatiotemporally synchronous pattern of spiking neurons; in addition, this kind of spike synchronization can emerge intermittently by turning information transmission delay between coupled neurons. Interestingly, with the advent of synaptic plasticity, spike synchronization is gradually destroyed by increasing synapse learning rate; moreover, the phenomenon of intermittent synchronization transitions becomes less and less obvious and it even disappears for relative larger learning rate. Further simulations confirm that spike synchronization as well as synchronization transitions is largely independent of network size. Meanwhile, we detect that large shortcuts probability can facilitate spike synchronization, but it is disadvantageous for delay-induced synchronization transitions.

In this paper, it is inevitable that the AR two-dimensional codes obtained by the mobile device will have a tendency towards the mobile augmented reality application. A two-dimensional code tilts correction method based on least square method is proposed. After binarization preprocessing, the target two-dimensional code inclination angle is obtained by least-squares method. After further correction by bilinear interpolation, satisfactory results can be achieved. Restore the image. The experimental results show that the method can be quickly and effectively identify the target two-dimensional code, and then the corresponding three-dimensional model to enhance the reality display.

In this paper, we introduce association rules into the event matching process, and propose a matching algorithm based on association rules in ontologybased publish/subscribe system. The algorithm discovers association rules from subscriptions, and then integrates these association rules into the arrived events. These integrated association rules guide the event matching process, which can greatly improve efficiency by reducing unnecessary matching processes. To solve the available problem of integrating the association rules into the ontology-based publish/subscribe system, we also show an approach of transforming the subscriptions into a format that can apply the data mining algorithm and an approach of integrating the association rule into the events. The results show that the algorithm achieves performance improvement of matching efficiency by using comparison.

Traditional collaboration systems are typically bundled with a set of tools with pre-specified static workflow and usage patterns. They are found to be inflexible and incompetent for today's heterogeneous and ever-changing collaboration contexts and computing environments. Today's collaboration systems must be aware of the dynamic changes of collaboration contexts in a timely manner and be adaptive to these changes appropriately. In this paper, we illustrate how business logic and computing environment context awareness are provided through a set of shared ontologies with well-defined semantics in a project called as EkSarva. The awareness of collaboration contexts and its changes then enables the collaboration system to adapt to these changes and be more flexible. A dynamic workflow process is also facilitated by the context-awareness.

Web services transaction handing is a key technique issue of computer supported corporate work. BTP protocol is a neutral protocol. It is not a specific web service transaction coordinating protocol. In order to enable BTP to support long term transaction, we made some modifications on it. These modifications include three aspects, which are, (1) To provide a strategy to eliminate negative effects caused by temporary results; (2) To put forward piecewise strategy of transaction node; (3) To put forward autonomy strategy of un-confirm collection elements. We also give design and implementation of transaction handling system, which is part of project “Model Driven Service Oriented Application Development and Integration Platform”, in which we use service broker method to extend a web service’s function, and enable services with transaction handling ability without making any modification on it.

Web service is a widely used technology, the current Web Service is mainly for non-state calculations. Because most business processes require long processing time, they need stateful calculations. This article introduces a method that combines workflow technology and Web Service technology, encapsulates workflows as business process services. We analyze different states of process services and different kinds of operational semantic impacts on the states, enable the process services can query workflow status at any time when interacting. And this method already applied in the reality, Stateful services have broad application prospect in the future.

This paper introduces the major process modeling methods and presents a new process modeling method based on Web service. The method enables business processes to be released and called as Web services and provides support for access service function. In the course of process execution, if an external service is required, the corresponding access service node will be added during process modeling. Traditionally, it is difficult for operators to take part in workflow system by Web service. To solve the problem, the author proposes a message management system in which the service provider and the requester communicate with each other via messages to realize manual participation.

Based on the Web, service customization is a new software development modality on the industrialized software integration development platform. The design and development of the customizable cooperative work framework?s architecture and software development tool?s running platform have been described in this thesis. The business object component?s attribute and presentation modality, service customization component and the business workflow instance can be customized through cooperative working of the frameworks to implement customizing the service process, which improves the effective of the manipulation among the business workflows and supports the flexible customization for the presentation user interface. As a new cooperative software development pattern, the customizable cooperative work framework can be used to customize a specific service process. The customizable cooperative work framework can support the customization and implementation of development and running the software archetype system in the special domain without recoding, re-compiling and re-deploying components.

Formal description of network behavior is the important part in network management. But some problems have not yet to be adequately addressed are the semantic and behavioral aspects of composition in this domain. Therefore a formal description method was presented to network behavior. The method includes a simple formal language, and its semantic was described by CSP. The formal language has graphical syntax with topology style. And it also has some characters that are suitable for the formal research and analysis about some problems in network management. Consequently, the method can be used to provide precise formal description and analysis about network behavior. And a reliable mathematical foundation is also provided for the correctness and validity of software system design.

In pervasive computing environment, collaboration between participants and interacts between applications are key characters of pervasive computing. In order to support those collaborative activities, we propose an ontology based workflow centric collaboration system modeling method. We build the collaboration system models in different abstraction layers according to domain related degree. Based on that, we can fulfill the requirements of variety and changeability of collaboration elements modeling. That enables the system with scalability and customizability.

With the rapid increasing of the number of services in Web environment, single or simple composition of the service is unable to provide high-quality service for requesters. Although dynamic service composition provides the automatic composition methods for machines, it cannot handle complex service requests. The main objective of researching Semantic Web is to extend the current World- Wide-Web, which makes all the information in Internet semantic. This paper arms to analyzes the main features of the Web Service. Then, with the definition and application of ontology, we propose the concept and definition format of service-ontology, and with the construction of service composition and corresponding algorithm, we apply it in the dynamic Web service composition application. Combined with intelligent smart transcript repository and ontology knowledge repository, the architecture provides the static and dynamic composition environments to facilitate knowledge accumulation, logic inference, binding web service and creation of process driven model. The approach considers the semantics of services along with the quality and the efficiency of the composition. Finally, based on the prototype system to validate the approach, the results of experiment prove that the Service Composition Engine is feasible and effective.

This article introduces a workflow centric collaboration system based on ontology, which is context-aware and adaptive. Using ontology to represent most collaboration elements and rules of the system, we introduce an ontology repository into the framework. It gives the system the capability to represent collaboration knowledge smoothly. The system focuses on business workflow transition and business data manipulation. It can be easily extended and upgraded by modifying the ontology repository when the requirements change. Based on ontology, the system can also interact with other ontology based systems and applications.

In pervasive computing environment it is required that applications should be able to discover and interoperate with each other. We designed and develop a workflow management system that is based on metadata. In the system we integrated document management, flow management and data management together. The flow engine of the system is capable of manipulating the business data when routing the flows based on metadata, and the system can access each other by interpreting opponent's metadata and context. The system provides visual metadata definition tools that help users to define specific application. That makes the system suitable to be used in various domains. When models of business flow or business data structure change, users can modify the metadata by using the tools to make their system adapt to the new requirements. In this way, the system can be agile and scalable.

This article proposes a collaborative workflow system based on ontology, which is focused on typical database access. By introducing ontology base and realizing the ontology definition of flow service, domain knowledge and user interface, the system is able to incorporate collaborative knowledge dynamically and thus is flexible enough to apply in other domains which have same application patterns. The extension and upgrade of the system can be achieved by updating and adding the definition of ontology, which is one of the distinct features of the system. With the ontology base, it is possible to enable the cooperation among different systems, the interaction among different workflow systems, as well as the interaction between workflow system and other applications.

Network security research is an important aspect of CSCW (computer supported cooperative work). It helps to make work environment essential and reliable. Access control security technology mainly includes firewall technology, intrusion detection technology, and security auditing technology. These technologies still have some problems and shortages though they are matured in some degree. The expert system for preventing and auditing intrusion is a series of software and hardware systems for reducing the risk of computer network security. It is an effective closed loop control system that integrates with firewall, intrusion detection, and auditing trail analysis. It uses the learning ability of expert system to add knowledge and rules. Also it combines the characteristics of real-time and unreal-time to form an access control security system and enhance the security of the network.

This paper presents our recent research on network performance measurement. From various perspectives, our work focuses on how to obtain the measuring data using the Simple Network Management Protocol (SNMP). Several models are proposed to measure the network performance effectively. These models are then evaluated and validated through implementations and analysis. The results show advantages of the proposed approach for network performance measurement based on SNMP and potential applications in telecommunication domain.