Search

Based on the activity for building the foundation and application of computer, a common used automatic examination paper generating system is designed. This paper introduces how to design the database of item bank and establish the mathematic model of examination paper generating system, and discusses the method of finding the solution. To solve the model, the multiple objectives are translated into single objective by using genetic algorithm.

Automatic test generating system in distributed computing context is one of the most important links in on-line evaluation system. Although the issue has been argued long since, there is not a perfect solution to it so far. This paper proposed an innovative approach to successfully addressing such issue by the seamless integration of an algorithm and multi-agent system. In the design phase, a test paper qualification model was defined for automatic generation of test paper with higher efficiency. The algorithm first generates a quantization model of test paper then the test items are selected from the database according to the quantization model. The random selection and approximate match tactics are used in this algorithm. In the phase of implementation, the algorithm was implemented in an agent by the Java Agent Development Environment. The final simulation results validated the feasibility of the proposed approach.

With the quickly development of the information system and the applications of six sigma in manufacturing company, six sigma faces the problem how to analyze the vast of data effectively. Data mining from simulation outputs is performed in this paper; it focuses on techniques for extracting knowledge from simulation outputs for a production and optimizing devices and labors with certain target. This paper first gives a brief definition of six sigma. Then we set up one simulation model for the production process and construct optimization objective. Then we set up one data mining model based on WITNESS Miner. It then explains six sigma project modeling with WITNESS incorporating the calculation of sigma ratings for processes and the export of key statistics. The WITNESS optimizer six sigma algorithm is explained and an example project illustrating the use of WITNESS Miner (data mining) is included. The mining results show that the model is able to fund important information affecting target, make manager diagnose the bottlenecks of the beer production process, and help manager to make decisions rapidly under uncertainty.

A range of issues often appear in current on-line courses selection system, such as systems crowded, excessive workload and so on. The paper seeks to construct a highly reliable courses selection system based on mainframe to address these issues. This paper boldly tried to combine mainframe technology with Web technology, use mainframe?s high reliability, high stability, and high concurrency of CICS?s on-line transaction to build the high reliability of courses selection system that can be allowed by a large number of students to select course online at the same time. The UI is designed of using Eclipse and MyEclipse plug-ins as platform, background by the IBM mainframe, DB2 technology, and CICS technology to complete the combination. The foreground Web interface is friendly and convenient, the background is running in mainframe CICS, stably and reliably. At present the system is running well, students have a good response and the senior school affirmed.

The complex software systems often involve many subsystems or components. Coordination among subsystems or components is a central issue in complex software development. The complexity of today software systems requires separation of coordinating entities from computational entities and runtime coordination. The paper presents a coordination framework that divides the system into coordinators and computational entities. The computational entities are represented as roles, processes and services so as to provide coordination with fine granularity. The coordinators are responsible for coordination among the roles. In our framework, coordinators are divided into three types: resource coordinator, communication coordinator and task coordinator. A set of general purposed data table I/O APIs are developed for relational databases, and some frequently used table I/O manipulations are provided as templates. In the paper, the strategies of coordination-oriented software development are discussed. A coordination framework based application development approach is proposed by a practical Commodity Trading System.

Microaggregation is an important technique to the k-anonymized datasets. However, most existing microaggregation algorithms to achieving k-anonymity have some defects on distance measurement for categorical and mixed data. In this paper, we introduce a categorical data semantic hierarchy to their distance measurement to improve clustering quality. The paper also investigates mixed distance for mixed data and designs an efficient microaggregation algorithm for them. Experiments show that the distance measurement for categorical data cause less distortion, and the improved microaggregation algorithm based on the mixed distance enjoys better clustering quality than the traditional MDAV algorithm.

Traditionally, drawing file transmission and graphic database fulfills Graphic sharing. With the emergence of Web as a publishing and programming medium, as well as the availability of high-level development tools, XML is used in graphic sharing to resolve the loss of data and file missing during the transmission. In this paper, extract the graphic information from DXF, which is widely supported format by most drawing design software. The combination of DLL and ArrayList techniques has been adopted to pick up graphics parameter. Graphic presentation can be built based on the properties of graphics entities; and by using the XSL and XML technique, it can be transformed into SVG, which is convenient to display on browsers. The paper offers an effective method for the sharing of graphic in the distributed environment.

The virtual electronic whiteboard is one of the main tools of the CSCL. This paper realizes an intelligent whiteboard system based on sketch recognition. The system is build on multitier distributed architecture, which can support the users to communicate with each other by sketching and chatting on pen based HCI. It can recognize the gestures to help users modify, copy and move sketches as familiar pencil-and-paper process, and transform the free-hand sketches into the symbols of the domain-specific knowledge automatically using feature points detection and graphics recognition algorithms. Such novel AI-aid sketching communication can support building connection between the concrete graphics and the abstract concepts during the learning activities, especially for information technology education. The experimental whiteboard system fulfills a flowchart-to-C programming and an ER diagram-to-database recognition and communication framework. The result of the usability evaluation of the experiment suggests that this system has higher efficiency for collaborative learning and more sufficient for the concept learning of information science.

The risk assessment of project is the important content for project management. This paper combines rough sets theory and support vector machine. The paper selects rough sets (RS) and support vector machine (SVM) algorithms to establish a new mathematical model for risk assessment of project. Using the rough sets to reduce numbers of indicators of risk factors, which reduces the dimensions of the input space. When treating the reduced data as the input space of SVM, we find that both the convergence speed and the assessment accuracy are enhanced. The results of Matlab simulation show the superiority of the model. The model based on rough sets and support vector machine can effectively help project managers for management of project risk.

A novel watermark algorithm is proposed for copyright protection of JPEG2000 compressed cultural heritage images in digital museum system. Visually meaningful color logo is encoded as the watermark information. Due to its characteristics, the key problem is to decide how to hide as many bits information as possible into the image while ensuring the information can be retrieved correctly, even after various attacks. In this paper, a new method is proposed to resolve it. For the frequency domain of color logo after Daubechies9/7 wavelet transform, just only low frequency and intermediate frequency coefficient is chosen as the watermark code. In order to ensure both the robust and transparency, intermediate frequency coefficient after Daubechies9/7 wavelet transform is also proposed to embedding watermark code. To achieve the lowest computational cost and provide security, the lifting scheme was presented to implement the wavelet transform. The central contribution of this paper is to develop a watermark algorithm, resilient to like JPEG2000 lossy compression attack and other mainstream attacks, embedding compressed color logo image as watermark and extracting it correctly. Experimental results show that the proposed watermark scheme is computational and quite robust in face of various attacks such as JPEG2000 compression with high compression ratios, filtering, cropping, etc.

The RobuCup 2D soccer simulation has been used as the basis for successful international competitions and research challenges and to simulate the interest of the public for robotics and artificial intelligence (AI). In cooperation strategy designing, researchers often employ the method of machine learning to optimize the simulation system, as the research growing, Q-learning algorithm, which is a particular type of machine learning, is becoming more popular. The paper is extended as follows: First, we will make a description of the characteristics about the RoboCup simulation system. Second, there is an analysis of the limitation about the traditonal Q-Learning algorithm, and a new strategy based on Q-Learning will be advanced here. Finally, Simulated results will be discussed and the paper comes to a conclusion.

With the development of telecommunication network, its large scale and complex framework caused the relationship among mass alarms of network elements to become more and more complex. It is necessary to provide network administrator with an approach to implement alarm filtering and fault location. To meet this need, an improved FP-Growth algorithm is given with which a new model is proposed based on mining association rules. Then the paper focuses on the alarm filtering and fault location with the help of these rules. An experiment to check the model and this algorithm is introduced and analyzed at the end of the paper.

Biomedical image processing is a relatively new but rapidly growing interdisciplinary area in both computer science and biology fields. This paper provides a systematic view and design method of the biomedical image processing course for undergraduate computer science major. The cross-discipline pedagogies that provide the guiding principles for course system design are presented, along with specific issues related to class room teaching and student practice. The paper also presents four improvements for the future course development.

With the ever-increasing application of embedded systems in the field of industrial control, the demand for embedded configuration software becomes inevitable. This paper presents a general method of developing embedded configuration software based on embedded Linux operation system. The software architecture consisting of host and target is given out. The relationship between graphic-element classes and its implementation are described. The animation implementation technology of dynamic graphic-elements and data communication mechanism based on signal and slot are discussed in detail. Furthermore, aimed at the poor processing power of embedded microprocessors and the limited storage resources on embedded platforms, the paper proposes a multi-level strategy for the performance optimization of embedded configuration software. The experiment result shows that this embedded configuration software can meet userpsilas requirements on real-time performance and resource consumption, and thus has application prospect and popularization value.

The main characteristic of the E-government system is the random and simultaneous happening of incidents, combining the strong ability of the Petri net description in the parallel system, this paper has proposed to use the Petri net theories to build up the E-government system. The paper has given an example in a carried out engineering item and proved soundness and performance analysis in the Petri net-based modeling.

For the different communications of specific NOC (network on chip) applications, this paper proposes a customized arbitral priority NOC router. This router uses the arbitral mechanism based on lottery algorithm instead of the RR (round robin) algorithm, which is widely used in the arbiter of current NOC routers. The arbitral priority of lottery router can be customized by users according to the communication cases among IPs in NOC. Through the analysis of different priority input ports competing for the same output port, it is illustrated that the lottery mechanism distinguish different priorities and guarantee the response to the high priority port. This paper uses SystemC to implement the router and build a NOC based on the 2D mesh to simulate and compare the performance of lottery router with RR router in average delay and the network resources utilization.

The Internet and the World Wide Web in particular provide a unique platform to connect learners with educational resources. Personalized curriculum sequencing is an important research issue for Web-based learning systems because no fixed learning paths will be appropriate for all learners. This paper propose a personalized recommendation approach for civil engineering curriculums based on collaborative filtering. This paper makes two critical contributions, firstly, it presents a civil curriculum item sequencing approach that will generate personalized curriculum item sequencing; secondly it illustrates the collaborative filtering to develop a summative examination or assessment analysis.

In this paper a teaching model has been established for compiler principle course through making engineering reform to compiler principles course. Ladder-like iterative teaching method has been introduced. Our teaching model and teaching method have provided solutions to the so-called three difficult problems, namely, the difficulties in teaching the compiler principles, the difficulties in understanding and mastering the course, and those in practical development. Moreover, the model and approach in this paper offer a theoretical basis for making engineering reform to other core curriculums of computer science & technology.

To realize the integration of different software systems, this paper introduces the agent technology and proposes one strategy, SCITR (scripting control integration strategy based on task-reduction method), that is used to achieve the agent system integration. Application of SCITR is based on the special agent structure and agent cooperation mechanism. Firstly, this paper respectively studies the current problems of the task-reduction method and the control integration technology, and then provides the demand for the integration rule that is described with the scripting language. Secondly, to the task-reduction method and the CISL (control integration scripting language) technology, this paper respectively studies their application in the realization of control integration. Finally, through applying SCITR to the Border and Coast Defense Simulation System, this paper studies its reliability and makes two contrasts with the traditional integration method RDS (repeat development strategy). The results show that SCITR could realize the agent cooperation and achieve the agent system integration with great efficiency.

Because of the fast growth of Web application and relative technology, the automatic testing for Web has become more and more complexity. This paper proposed an automatic system for Web-app functional testing which based on a XML description and also integrated the web pagespsila structure analysis. It improved the traditional way of record, and it can auto-generate test data and test cases by the constraints and the rules we defined in this paper. This system can record the path that user operated on a Web with all the pages passed by. Referring the definition of workflow, it defines a test-flow and described in XML. It can do test work as a user when flowing the test-flow with using the data which was been generated, so it has implemented a more automatic testing.

"Parallel development and independent integration software development model" is proposed in this paper, which takes the characteristics of software development in scientific research field into consideration, absorbs fully the merits of available software development model, and makes good use of some recent thoughts and technologies such as software reuse, components, UML and so on. Each part of the model including the definition, the assignment and the end products is analyzed and described in detail in this paper, and finally, the validity and practicability of this model is proved in practice.

The paper introduces a data security assurance mechanism, which is already applied in a security data exchange platform (SDEP). Then, the paper describes the security design of SDEP in details. By using the security solution, information sender and receiver can build a security communication channel, and transfer lots of private files.

This paper shows how event-driven software architecture may facilitate easy debugging. Most existing replay-based debugging techniques involve tool developments. Drawbacks for such techniques are two-fold: (1) extra cost for tool development. (2) lack of availability, tools and techniques for one platform might not be available or applicable on other platforms. This paper tries to improve the situation by facilitating replay at software architecture level. Programs conforming to the proposed architecture are replayable by construction. There is no need for tool development and the technique is available as long as the used programming language is available on the target platform. Two notions, namely pure property and pure entity, are proposed. It is proved that bugs in pure entities and pure properties can be systematically found with the support of the underlying replay-oriented architecture. The architecture is applied to the development of a typical distributed system, which is used to manage a collection of directory servers. Experience with this application shows that more than half of user developed code can be characterized as pure entities and pure properties. Our experience also shows that a large portion of impure properties can still be debugged using this architecture. It is also explained why this replay-oriented software architecture is generally applicable to other applications.

Spatial datum acquisition is one of fundamental components and main datum source of most GIS application systems. As a result, providing a universal framework for datum acquisition to satisfy the needs of those professional GIS applications is significant in GIS applications. The framework presented in this paper changes the traditional ways of datum acquisition, thus improving the efficiency greatly and reducing most potential man-made mistakes. Moreover, this paper also addressed those key algorithms employed in the framework and supplied their implementations.

Assembly language programming courses inevitably include the need to assess programs submitted by the students. However, the task of grading solutions to student programming exercises is laborious. This paper describes a system for the automatic assessment of assembly language programs. This assessment tool uses the MASM, a developing tool for assembly language programs. This paper explains the background and aims for developing this system. It focuses on the assessing strategy and implementation.

With different kinds of terminal clients coming up, adaptive content delivery is a hot research topic. The proxy, or the intermediary server, is often chosen to do the function of producing ldquoadaptiverdquo contents in real time. Admittedly, those resolutions addressed the adaptive aspect of content delivery very well. However, they did not make full use of another proxy's important character, which is caching. In this paper,we concern mainly about caching the adaptive content. The object in our consideration is JPEG2000. The proxy in our paper support storing partial object in the cache and realize adaptive increment of the quality of copy in the cache for the subsequent requests. After testing the object perception time, we conclude that the performance can be enhanced after making partial object caching supported.

To solve the problem that abnormal traffic including Internet worm and P2P downloading has occupied the LANpsilas bandwidth, a danger-theory-based model to detect anomaly traffic in LAN is presented in this paper. The definition is given, in this paper, to such terms as dangerous signal, antigens, antibodies and memory antibodies. Besides, matching rule between antigen and antibody is improved. Experiments show the outstanding performance of the proposed model in real-time property, high detection rate and unsupervised learning.

Grid resource management and scheduling are confronted with a great challenge. This paper presented a dynamic resource price-adjusting (RPA) strategy in computational Grid. Three new concepts and two new evaluation standards are defined. On the basis of these theories, a Grid resource scheduling algorithm based on marker-driven using dynamic RPA strategy is proposed. With the help of GridSim toolkit, the simulation of resource scheduling algorithm proposed in the paper is realized, and the new algorithm is compared with GridSim DBC (deadline and budget constrained) scheduling algorithms. The experimental results show that the new algorithm improves the performance of Grid in aspect of increasing the task accomplishment ratio and solving the problem of load imbalance.

Auto tree classification is of use in plant research. We put forward and implement a hibernal tree automatic classification system. In the paper features of tree system and other aspects that could influence classification results are analyzed and taken into consideration in classifier modeling. We extract the most related contents and information for classification, set up hibernal trees classification model, and finally accomplish a hibernal tree automatic classification system based on Bayes. We also make use of Bayes Network in coefficient learning so as to get best classification effects by adaptive self-learning and necessarily adjusting parameters according to actual data. Experiment result shows the method proposed in the paper can well solve hibernal tree automatic classification problems.

This paper presents a method for test case selection that allows a formal approach to testing software. The two main ideas are that testers create stochastic models of software behavior instead of crafting individual test cases and that specific test cases are generated from the data flow-based models and applied to the software under test. This paper describes a method for creating a data flow-based model in the context of a solved example. We concentrate on Markov models and show how non-Markovian behavior can be embedded in such models without violating the Markov property.

Generally, enterprises' production-run parameters will affect the height of the energy consumption. This paper using the methods of factor analysis and fuzzy association rule mining to evaluate the height of enterprises' energy consumption, and bring forward an optimization method to improve the high energy consumption parameters, a power plant case is presented at the end of the paper to illustrate the proposed model.

With the growth of publishing scientific literature on the World Wide Web, there is a great demand for clustering online scientific literature by using the citation patterns. A scientific community in the citation graph represents related papers on a single topic. In this paper we improve the random walk graph clustering algorithm to find scientific communities by using correlation coefficient in the citation graph as the similarity metric. Our experiment results show that the approach performs better than the original random walk graph clustering method.

This paper introduces a method about the improvement of JPEG2000 in retaining texture of image by using the transform of bandelet. The contours and textures of image are usually the important elements, especially in remote sensing image. However these textures are lost when JPEG2000 is used in the condition of large compression ratio. In this paper, bandelet is used to improve JPEG2000 because of its compatibility with JPEG2000. Wavelet coefficients are partitioned to a lot of code-blocks in JPEG2000. And these code-blocks naturally act as the units of the bandelet transform in this paper. And then some tricks are applied in the follow steps. Experiments show that with modifications mentioned above, more detail information is retained compared to JPEG2000. In addition, the peak signal noise ratio (PSNR) of image gains a little by using bandelet in compression process.

Advances in computing applications in recent years have prompted the demand for more flexible scheduling models. The DWCS (Dynamic Window-Constrained Scheduling) is a flexible scheduling algorithm. This paper applies VDS (Virtual Deadline Scheduling) and DWCS into the open real-time system. The paper analyzes the CBS (Constant Bandwidth Server) which the open real-time scheduling is based on. To evaluate the performance of the new server, the two server mechanisms have been implemented on the Linux kernel. The experimental results show that the new server is better than the old server. In the end, the future research directions are discussed.

To solve the problem how to use different business rule engineer in an identical system, a combinatorial model of BRE was put forward in this paper. In this model, this paper achieved loosely coupled integration of a few engineers in a service-oriented way, and relevant theory was applied to analyze the process. Finally, study of a case explained the feasibility of this model.

In the data integration, the heterogeneity of sources leads to missing value and various expressions of the same value in records, which reduces the quality of data. In this paper, we propose a novel approach to reinforce records for the integrated data. By studying the functional dependency of attribute in schema of data integration, we discover the related attribute that determines the attribute with uncertain value. Then our approach exploits matching algorithms on the value of related attribute to associate different records. And the uncertain value will be reinforced with the consistent value in a certain record. We also propose algorithms for reinforcing dataset of data integration. The experiments based on the data of conference paper demonstrate the effectiveness and performance of our approach on improving the quality of data.

This paper proposes an SVG-based spatial analysis method to support spatial query and analysis in the Web environment. We discuss two issues of this method. First, based on the theory of spatial data modeling, we propose an SVG-based spatial information representation model to represent the needed geospatial data in SVG. Furthermore, this paper designs and implements some spatial operators and integrates them into an SVG-based spatial extended SQL to support spatial analysis on spatial data represented in SVG. This method can improve the functions of current WebGIS applications, most of which have been only employed for visualization. Finally, this paper implements the method and designs a case study to evaluate the proposed method.

A new vectorization method of dot matrix image based on error controlling is studied in this paper. The least-square algorithm is a main method of vectorization of dot matrix image. In the calculation of the fitting error by using the least-square algorithm, if the fitting error is in the range of threshold value, the result of vectorization will be considered to be right. So how to determine the threshold value is the key. In this paper, it has analyzed the vectorization errors of ideal bitmap by least-square algorithm and put forward the judgment formulas to evaluate recognition result. In order to complete vectorization automatically, it has presented a new method of the vectorization of dot matrix image-the least-square rolling recognition algorithm for double direction. It has solved the problem of selecting suitable dot group. The judgment formulas in this paper are scientific and accurate; the method of the vectorization is rapid and high efficient.

In order to build an open, collaborative and efficient visual group learning system (VGLS for short below) in a wide area network (WAN for short below)environment, such as the Internet, this paper studies how to combine multiple techniques such as the mobile agent, ontology and Web services composition to build a VGLS. As the research result, a distributed model for VGLS (DM4VGLS for short below) is proposed in this paper. The application framework of the DM4VGLS model is illustrated. Based on the above application framework, the components within the DM4VGLS are introduced with an emphasis on the collaborative mechanism of them. A simulation prototype based on jadex, a belief desire intention (BDI) reasoning engine that allows for programming intelligent software agents in XML and Java, is developed. The initial experiments show the feasibility of the proposed DM4VGLS model.

The process of traditional requirements acquisition and analysis is generally completed by domain experts and requirements analysts, which will greatly restrict the requirements modeling automatization degree and userpsilas participation degree in the requirements process. This paper presents a reflective mechanism for evolution of requirements specifications (RMERS) that supports OWL-S. In the phase of requirements analysis, OWL-S is used to describe system requirements specifications, OWL-SA is used to describe requirements change, the evolution of requirements specifications caused by userspsila individualized requirements is completed via reflective mechanism. Besides, this paper also explains the description language OWL-SR of reflective requirements specifications of this reflective mechanism and the meta-information model. By means of this reflective mechanism, through evolution of OWL-S requirements specifications, the task of requirements analysis of networked software can be completed with high efficiency in the phase of requirements analysis.

Traditional support vector machine needs pre-assumed kernel functions. This paper proposes a method via gene expression programming to automatically construct the kernel. The contributions of this paper include: (1) proposing the concepts of GEP kernel and kernel tree; (2) proposing the properties of GEP kernel and the kernel relation theorem; (3) proposing GEP based support vector machine (KGEP-SVM), (4) decoding kernel individual algorithm (DKIA) and kernel operators operating algorithm (KOOA), and (5) extensive experiments show that the average accuracy of the method is increased by 4% and generation of GEP kernel is about 150.

With the increase of the popularity of Web service, more and more Web applications are developed with this new kind of components. Web service quality validation and control is becoming very critical to vendors, brokers and application builders. Both brokers and users usually have to use black-box testing because design and implementation details of Web service are unavailable. This paper proposes a formal approach for automated test data generation for Web service single operation based on WSDL specification. Firstly, we define a formal model for the datatype of input element of the operation. Secondly, the paper presents an algorithm that derives the model from WSDL. Thirdly, the paper presents a method that generates test data for operations from the model. Examples of using this approach are given in order to give evidence of its usefulness.

A new method of fault diagnosis based on feature weighted FCM is presented. Feature-weight assigned to a feature indicates the importance of the feature. This paper shows that an appropriate assignment of feature-weight can improve the performance of fuzzy c-means clustering. Feature evaluation based on class separability criterion is discussed in this paper. Experiment shows that the algorithm is able to reliably recognize not only different fault categories but also fault severities. Therefore, it is a promising approach to fault diagnosis of rotating machinery.

In order to improve the reusability for the software testing automation procedures, at the same time support the program test for Linux, keyword-driven distributed test automation framework (LKDT) is proposed in this paper, which is based on the analyze of the recent automated testing frameworks and combined with the actual project practice for software testing automation. This paper presents the LKDTpsilas system architecture, design, as well as the key technology details, finally, a concrete application and instance analysis are given to estimate the frameworkpsilas feasibility and superiority.

Pattern recognition has been successfully used in different application areas, its application on identifying playerpsilas strategy during the gameplay which is called player strategy pattern recognition (PSPR), is another interesting area. PSPR can greatly improve game AIpsilas adaptability, and as a result the entertainment of game is promoted. In this paper, pac-man game is used as a test-bed. Kernel classifier of radial basis function (RBF) is chosen to analyze off-line data from gamers who are choosing different strategies, in other words the classifiers are trained with sample data from players using different strategies. The method attempts to use the trained classifier to predict strategy pattern of a future player based on the data captured from its gameplay. This paper presents the basic principle of the PSPR by using the RBF theoretic approach and discusses the results of the experiments.

In order to increase transfer rate and enhance performance in mass data transmission and storage in real time system, a high performance Serial Advanced Technology Attachment generation 2(SATA II) host controller is proposed in this paper. This paper presents how to improve the hardware performance in course of implementation and validation. The controller was designed using VHDL and developed in the Xilinx Integrated Software Environment (ISE) Version 9.1i. Link layer and transport layer of the host controller architecture are realized in Xilinx Virtex-4 FPGA. SATA protocol is realized based on Multi-Gigabit Transceiver (MGT). Chipscope is applied to fully test and verify the design. Comparing with the routine design method, the new project has better features in the expansibility, scalability, improvability and in-system programmability. The experiment results show that the controller works accurately and stably at 3 Gb/s transfer rate. The controller is well suited for high speed data transfers.

With the development of the Internet, the need of city 3D data is rising more and more, and the emphasis shifts from analysis to the sharing of city 3D data and information over the Internet. In Internet, we lack uniform 3D data standard and 3D data cannot be shared and exchanged availably. The city Geography Mark-Up Language (GML) was developed as the standard language for exchanging and sharing city 3D data in Internet. In this paper, we propose a set of data structures to obtain city 3D information from the GML file in order to extract useful information quickly, after analyzing CityGML files structure and parsing by the Document Object Model (DOM). The experiment shows that 3D data are extracted effectively, and visualized in Web application using Java, and the data structures defined in the paper are very validated.

Digital Education Service (DES) or Digital Education Service Chain (DESC) is also a brand-new field of the modern service industry. This paper aims to prepare theoretical foundation and technological basis for DES and DESC. In this paper, the concept of DESC is discussed. From the perspective of service chain, a variety of DESCpsilas models are given and analyzed in figure and formal mode. According to SaaS (Software-as-a-Service), the service architecture of DESC is figured out. At the same time, the IT architecture for DESC is designed based on SOA. Finally, the further research directions are put forward.

Web-based testing is the trend of the modern educational testing. But how to upload the formula in the testing item easily is a difficult problem. At first, this paper formulates the main comprehension of PIA and the uploading strategy based on PIA. After that, as an example of its application, this paper introduces a Web-based mathematics testing system.

The Mobile Agent (MA) has been seen as a promising distributed computing technology. According to the analysis of security problems of current mobile agents that are exposed in opening environment, a model system for mobile agent is proposed in this paper. In this system, the mobile agent security is enhanced by the Trusted Computing Technology. The mobile agent security supporting environment (MASSE) is designed by using trusted software stack (TSS) and trusted computing platform (TCP). To improve the security of MA system, some MA security services, such as authentication, role based access control and MA negotiation, are discussed in this paper. The working process of MASSE is also described.