Innovation networks provide an efficient mechanism for organizations to realize their potential for learning and innovation improvement. Firms situated within innovation networks require specific abilities to acquire, from their network partners, the knowledge and the complementary assets that facilitate their innovation performance. Drawing on the resource-based view and social network theory, this study identifies two types of network capabilities: network structural capability and network relational capability. The purpose of this study is to deepen our understanding of the precise manner in which these network capabilities affect the networked firm's innovation performance. Based on the data obtained from Chinese high-tech firms, this study's findings suggest that network structural capability has a greater positive impact on innovation performance than network relational capability does within an exploration-orientated network. However, network relational capability is more positively associated with innovation performance within an exploitation-orientated network.
Cyberspace is the new geographical space formed in the virtual digital world which is different from the traditional geographical system. There are some debates that whether Cyberspace lead to further spatial divergence of economic activities or not. Based on existing related research achievements this paper studies the Cyberspace's influence on the spatial distribution of business services, puts forward some hypothesizes about the relationship between the Cyberspace active location and the agglomeration of business services enterprises. By some empirical method, it measures the spatial coupling of location's activity in Cyberspace and the business service enterprises' agglomeration, and the econometric model is established to test the hypotheses. It finds that there is a high coupling relationship of location's Cyberspace activity degree and spatial agglomeration of business services.
We present a metaheuristic algorithm for testing software, especially web applications, that can be modelled as a state transition diagram. We formulate the testing problem as an optimization problem and use a genetic algorithm to generate test cases as sequences of events. This algorithm evolves solutions by maximizing a fitness function that is based on testing objectives such as the coverage of events, diversity of events, and continuity of events. The proposed approach includes weights that can be assigned to events. These events would lead to important features or web pages in order to ensure that test cases will be generated to cover these features. The effectiveness of the genetic algorithm is compared with that of other algorithms, namely simulated annealing and a greedy algorithm. Our experimental results show that the proposed genetic algorithm demonstrates serious promise for testing state-based software, especially web applications.
SDN(Software-Defined Network) is one of the hottest networks under research and development. By comparing with policy based network and softswitch, the principles of OpenFlow and SDN are introduced, and the SDN architecture is described. Its innovation in the network programmability and virtualization are illustrated in detail. The main problems of SDN such as application software license model, global status information collection and central controller scalability in large scale network are analyzed. The typical application scenarios in local autonomous networks are discussed, with the cross-layer integration of IP, optical and wireless networks emphasized, and the instance of Data Center virtual networking supporting Cloud Computing is presented. As an example of SDN application in autonomous WAN, networking of the multi-domain UC system is studied in detail, in order to address unsolved issues such as overlay management overhead and bandwidth costliness. After describing the architecture of multi-domain UC system, the typical transmission networking structure is designed with an overlay between provincial domains based on SDN, and the bandwidth requirements of signaling and media traffics are estimated. By comparing the performances of overlays with mesh, star and hybrid topologies, the hybrid overlay topology is showed to be the best choice in terms of cost and manageability for the typical traffic distribution, with bandwidth utilization up to 84%. However, the requirements on user technology capabilities will be too much.
We develop an Android application "Flying Dictionary" which is an offline pop-up dictionary meant to ease users by displaying the meaning of the words they find tough without having to switch to some other app in order to look for its meaning. The application was created and uploaded on Play Store in March 2017, and since then over 4000 users have installed it on their Android devices. With a current rating of 4.8 (as of December 2017), our aim is to make it easier for otherwise reluctant students to encourage learning new words with ease and at the click of a button. This paper analyses the impact of using the application for daily look-up of meanings, on users' learning experience. The analysis indicates that providing the users with the ability to glance the meanings on the go helps them in learning more new words and prevents them from giving excuses like the need for an online connection or the need for switching the app in order to look for the meaning. This contributes to the growing importance of the pop-up feature and emphasizes the need to incorporate this in the popular dictionary apps as well as try and bring up the feature as a built-in utility in the Android OS itself.
This paper proposed a new method on the combined synchronization method for different structure time-delayed fractional-order systems, which contain uncertain parameters and disturbs. Based on the Lyapunov-Krasovskii stability theory, the combined synchronization controller is designed between two drive system of time-delayed Rossler system and Lorenz system and one response system of time-delayed Liu system, which is taken as an example. Numerical simulation results show the effectiveness and validation of the proposed synchronization scheme.
In this paper, we present the analytical development of a new fuzzy strategy applied to a fire control station. An Intelligent Fire Control Central Station (IFCCS),or Intelligent Fire Alarm Central Station (IFACS), is the controlling component of a Fire Alarm System. The panel receives information from environmental sensors designed to detect changes associated with fire, monitors their operational integrity and provides for automatic control of equipment, and transmission of information necessary to prepare the facility for fire based on a predetermined sequence. The panel may also supply electrical energy to operate any associated sensor, control, transmitter, or relay.There are four basic types of panels: coded panels,conventional panels, addressable panels, and multiplex systems.
Utilizing the driving characteristics of the hybrid electric bus with UC facilitate to achieve better fuel efficiency than that without UC. We proposed the standards of when and how much to use the UC with regard to the acceleration and SOC such that much improved vehicle without changing the structure could be developed. The fuel economy was tested using the Thermostat Control Strategy and the fussy logic controller was used to define the operation limit of the UC such that the significance of this research could be verified.
We develop a new approach for distributed computing of the association rules of high condence in a binary table. It is derived from the D-basis algorithm , which is performed on multiple sub-tables of a given table. The set of rules is then aggregated using the same approach as the D-basis is retrieved from a larger set of implications. This allows to obtain a basis of association rules of high condence, which can be used for ranking all attributes of the table with respect to a given xed attribute using the relevance relevance parameter introduced in . This paper focuses on the technical implementation of the new algorithm.Some testing results are performed on transaction data and medical data.
In recent years, deep learning has made leaps in the fields of artificial intelligence, machine learning and so on, especially in the fields of speech recognition, image recognition and self-learning. The deep neural network is similar to the biological neural network, so it has the ability of high efficiency and accurate extraction of the deep hidden features of information, and can learn multiple layers of abstract features, and can learn more about Cross-domain, multi-source and heterogeneous content information. This paper presents an extraction feature based on multi-user-project combined depth neural network, self-learning and other advantages to achieve the model of personalized information, the model through the input multi-source heterogeneous data characteristics of in-depth neural network learning, extraction fusion collaborative filtering widely personalized generation candidate sets, and then through two of models to learn to produce a sort set, Then realize accurate, real-time, personalized recommendation. The experimental results show that the model can study and extract the user's implicit feature well, and can solve the problems of sparse and new items of traditional recommendation system to some extent, and realize more accurate, real-time and personalized recommendation.
Literature review is part of scientific research. Online references management tools help researchers in finding relevantliterature and documents. Finding relevant conferences isthe key step to understand the research field. Researchersusually rely on the conference names to find out whetherthey are related. However, the conference name rarely reflects the diverse topics it covers. For instance, for thetwo conferences, "International Conference on Data Miningand Applications" and "Special Interest Group on Information Retrieval" which represent similar research topics and research areas, but thenames fail to capture the similarity. One possible method tocompute the similarity between all the papers in the two conferences but it's time-consuming. Instead of computing thesimilarity, this work builds a search engine based on Luceneand find similar conferences given a query conference basedon the index. A BFS-based algorithm is proposed to addressthis problem and experiments on DBLP dataset shows theproposed approach can generate comparable results with thesimilarity-based approach
Software-defined networking (SDN) is reshaping the networking paradigm. Previous research shows that SDN has advantages over traditional networks because it separates the control and data plane, leading to greater flexibility through network automation and programmability. Small business networks require flexibility, like service provider networks, to scale, deploy, and self-heal network infrastructure that comprises of cloud operating systems, virtual machines, containers, vendor networking equipment, and virtual network functions (VNFs); however, as SDN evolves in industry, there has been limited research to develop an SDN architecture to fulfil the requirements of small business networks. This research proposes a network architecture that can abstract, orchestrate, and scale configurations based on small business network requirements. Our results show that the proposed architecture provides enhanced network management and operations when combined with the network orchestration application (NetOApp) developed in this research. The NetO-App orchestrates network policies, automates configuration changes, and manages internal and external communication between the campus networking infrastructure.
This research work was designed to utilize multi-level cyber crime detection and control system to provide enhanced real-time evidence to cyber crime investigators to aid them in prosecuting cyber criminals. The design was based on a robust system combining user-identity, device identity, geographical location and user activities to provide evidences to uniquely identify a cyber user and detect crimes committed. The system captures the user's facial image and biometric finger print as mandatory login parameters in addition to username and password before granting access. The system was tested and implemented in an real time cyber security website www.ganamos.org. The results showed that it is possible to divulge the identity of cyber users and associate their activities with the devices they use, the date, time and location of operation. These can provide real-time evidences to law enforcement agencies to track down and prosecute cyber criminals.
Copyright © CSITY 2018