The existing cross-level High Utility Itemsets Mining （HUIM） algorithms consume a lot of time and occupy large amounts of memory. To address these problems， a Data Index Structure Cross-level High utility itemsets mining （DISCH） algorithm was proposed. Firstly， the utility list with taxonomic information and index information was expanded into Data Index Structure （DIS） to efficiently store and quickly retrieve all itemsets in the search space. Then， in order to improve the memory utilization， the memory occupied by the utility lists that do not meet the conditions was reclaimed and reallocated. Finally， during the construction of utility list， early termination strategy was used to reduce the generation of utility list. Experimental results based on real retail datasets and synthetic datasets show that compared with the CLH-Miner （Cross-Level High utility itemsets Miner） algorithm， DISCH reduces the running time by 77.6% on average and the memory consumption by 73.3% on average. Therefore， the proposed algorithm can search the cross-level high utility itemsets efficiently and reduce the memory consumption of algorithm.

For the requirements of building Project Information Portal (PIP) data center based on a unified data model, a manageable entity object-oriented data model was proposed. The project data were treated as a series of managerial entity based on management workflows which were decomposed according to the whole life cycle. The conceptual layer data model was designed. The project data could be naturally represented and recorded by using this model. The data organization method was presented based on MongoDB (document-oriented database technology). The cluster storage architecture for PIP was also addressed. The experiments show that it has efficient performance in data writing and querying. It also has high availability and storage capacity scalability.

This paper focused on the environmental characteristics of High-Speed Railway (HSR) and its influence on 3rd Generation (3G) mobile communication networks. Firstly, the technical characteristics of three major 3G communication standards in China were analyzed, especially the two key issues in the construction of HSR communication system: wireless coverage and fast handoff. Then the body loss of serials of HSR, the minimum switching area and the optimization deployment strategies of 3G communication technologies in China were summarized. Lastly, the tendency of HSR's wireless communication technologies was prospected and it was found that the Long Term Evolution (LTE) technologies could provide the technical feasibility for the broadband application services of HSR. Hence, it is suggested that we should strengthen the research for wireless transmission performance optimization of TD-LTE in HSR environment.

This paper presented a sophisticated power saving scheme based on system-level resource management for TH-1A supercomputer system. The scheme introduced a uniform framework for centralized management of various power-consuming resources, i.e. computing elements, communication components, power supply and cooling devices. And many efficient management policies such as LRU etc. were applied within the framework.

The amount of test data for integration of Web services is quite huge, which influences the quality and efficiency of testing. A method of test data selection for Web services composition based on mutation technology was proposed. The paths were generated from BPEL (Business Process Execution Language) document. Then interface mutation operator and path mutation operator were used to generate mutants. The effective test data was selected according to the ability of the initial test data that kills the mutants. The experimental results show that the method is effective and the selected test data are less but more effective.

Aiming at the current de-duplication algorithms, two repeated sequences (RS)extracting algorithms were compared and analyzed. Since STC has favorable performance in considering time cost and the inverted index method is superior in terms of spatial complexity, STC was used to improve RS algorithm. Experiment results show that this method can find similar Web pages efficiently. This algorithm can reach a high precision in mono-language deletion of duplicated Web pages, and this algorithm can also reach a maximum precision when it is applied to deletion of duplicated web pages.

Through studying multi-type arithmetic in many systems, it is found that the existing solutions are not flexible. Process-oriented solution and two potential object-oriented solutions were given and their shortages were analyzed. To achieve flexibility, reusability, maintainability and efficiency, a two-layer virtual table object-oriented solution was proposed based on agile design policy and the Open-Closed Principle (OCP). Analysis shows that the solution is relatively good.

A feasible way to stimulate cooperation in selfish Ad Hoc networks is to reimburse the forwarding service providers for their cost. We proposed a low overhead truthful energyefficient protocol called LOTER, which introduced the VCG mechanism into the selfish wireless Ad Hoc networks. In LOTER, the node collected its neighbors' information. The destination node selected several RREP paths. If it was on the RREP path, the node sent its collected information to the source node. Based on such information, the source node found the least cost path to the destination and calculated the VCG payment to the nodes on the chosen least cost path. LOTER can realize truthful and energyefficient routing in the selfish Ad Hoc networks. Furthermore, it is skillful at collecting topologic information with low overhead. Stimulation results show that LOTER provides better performance than other existing protocols.

Based on the approximation theory of Fourier-series working in square integrable space, a Fourier neuronal network was constructed by using activation functions of the complex exponential form. Then a weightsdirectdetermination method was derived to decide the neuralnetwork weights immediately, which remedied the weaknesses of conventional BP neural networks such as small convergence rate, easily converging to local minimum and possibly lengthy or oscillatory learning process. A hidden-neurons-growing algorithm was presented to adjust the neural-network structure adaptively. Theoretical analysis and simulation results substantiate further that the presented Fourier neural network and algorithm could have good properties of high-precision learning, noise-suppressing and discontinuous-function approximating.