To address the problems of limited information expression， imbalance， and dynamic spatio-temporal characteristics of accident data， an accident prediction model fusing heterogeneous traffic situations was proposed. In which， the semantic enhancement was completed by the spatio-temporal state aggregation module through traffic events and weather features representing dynamic traffic situations， and the historical multi-period spatio-temporal states of four types of regions （single region， adjacent region， similar region， and global region） were aggregated； the dynamic local and global spatio-temporal characteristics of accident data were captured by the spatio-temporal relation capture module from both micro- and macro-perspectives； and the multi-region and multi-angle spatio-temporal states were further fused by the spatio-temporal data fusion module， and the accident prediction task in the next period was realized. Experimental results on five city datasets of US-Accident demonstrate that the average F1-scores of the proposed model for accident， non-accident， and weighted average samples are 85.6%， 86.4%， and 86.6% respectively， which are improved by 14.4%， 5.6%， and 9.3% in the three metrics compared to the traditional Feedforward Neural Network （FNN）， indicating that the proposed model can effectively suppresses the influence of accident data imbalance on experimental results. Constructing an efficient accident prediction model helps to analyze the safety situation of road traffic， reduce the occurrence of traffic accidents and improve the traffic safety.

In order to solve the problem that it is difficult to obtain a composite service with high overall performance in a large-scale Web service environment， a large-scale Web service composition method was proposed. Firstly， Document Object Model （DOM） was used to parse the user demand document in XML format to generate an abstract Web service composition sequence. Secondly， the service topic model was used for service filtering， and Top-k specific Web services were selected for each abstract Web service to reduce the composition space. Thirdly， in order to improve the quality and efficiency of service composition， an Optimized Grey Wolf Optimizer based on Logistic chaotic map and Nonlinear convergence factor （OGWO/LN） was proposed to select the optimal service composition plan. In this algorithm， chaotic map was used to generate the initial population for increasing the diversity of service composition plans and avoiding multiple local optimizations. At the same time， a nonlinear convergence factor was proposed to improve the optimization performance of the algorithm by adjusting the algorithm search ability. Finally， OGWO/LN was realized in a parallel way by MapReduce framework. Experimental results on real datasets show that compared with algorithms such as IFOA4WSC （Improved Fruit Fly Optimization Algorithm for Web Service Composition）， MR-IDPSO （MapReduce based on Improved Discrete Particle Swarm Optimization） and MR-GA （MapReduce based on Genetic Algorithm）， the proposed algorithm has the average fitness value increased by 8.69%， 7.94% and 12.25% respectively， and has better optimization performance and stability in solving the problem of large-scale Web service composition.

Focusing on the shortages of easily falling into local optimal, low convergence accuracy and slow convergence speed in Particle Swarm Optimization (PSO) algorithm, an improved Particle Swarm Optimization based on HierarChical autonomous learning (HCPSO) algorithm was proposed. Firstly, according to the particle fitness value and the number of iterations, the population was dynamically divided into three different classes. Then, according to characteristics of different classes of particles, local learning model, standard learning model and global learning model were respectively adopted to increase particle diversity and reflect the effect of individual difference cognition on performance of algorithm and improve the convergence speed and convergence precision of algorithm. Finally, HCPSO algorithm was compared with PSO algorithm, Self-adaptive Multi-Swarm PSO algorithm (PSO-SMS) and Dynamic Multi-Swarm PSO (DMS-PSO) algorithm on 6 typical test functions respectively. The simulation results show that the convergence speed and convergence accuracy of HCPSO algorithm are obviously higher than these of the given algorithms, and the execution time difference of the proposed algorithm and basic PSO algorithm is within 0.001 orders of magnitude. The performance of the proposed algorithm is improved without increasing complexity.

Focused on the issue that interference would exist in the analysis and prediction of airport energy data because of the outliers, a prediction model based on improved Fuzzy Support Vector Regression (FSVR) was established for the demand of airport energy. Firstly, a fuzzy statistical method was selected to make an analysis on test sample sets, parameters and the outputs of models, and a basic membership function form consistent with the data distribution would be derived from this analysis. Secondly, relearning of membership function would be performed with respect to expert experiences, then the parameter values a and b of the normal membership function, the boundary parameter values of semi-trapezoid membership function and the parameter values p and d of triangular membership function would gradually be refined and improved, so as to eliminate or reduce the outliers which were not conducive to data mining and reserved the key points. Finally, combined with Support Vector Regression (SVR) algorithm, a prediction model was established and its feasibility was verified subsequently. The experimental result shows that, compared with Back Propagation (BP) neural network, the prediction accuracy of the FSVR increases 2.66% and the recognition rate of outliers increases 3.72%.

Spatial clustering is one of the vital research directions in spatial data mining and knowledge discovery. However, constrained by the complex distribution of uneven density, various shapes and multi-bridge connection of points, most clustering algorithms based on distance or density cannot identify high aggregative point sets efficiently and effectively. A point clustering method based on spatial proximity was proposed. According to the structure of point Voronoi diagram, adjacent relationships among points were recognized. The similarity criteria was defined by region of Voronoi, a tree structure was built to recognize point-target clusters. The comparison experiments were conducted on the proposed algorithm, K-means algorithm and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. Results show that the proposed algorithm is capable for identifying clusters in arbitrary shapes, with different densities and connected only at bridges or chains, meanwhile also suitable for aggregative pattern recognition in heterogeneous space.

Since the rapidly increasing data scale imposes a great computational challenge to the problem of Sparse Subspace Clustering (SSC), the existing optimization algorithms e.g. ADMM (Alternating Direction Method of Multipliers) for SSC are implemented in a sequential way which is unable to make use of multi-core processors to improve computational efficiency. To address this issue, a parallel SSC based on coordinate descent was proposed,inspired by a simple observation that the SSC can be formulated as a sequence of sample based sparse self-expression sub-problems. The proposed algorithm solves individual sub-problems by using a coordinate descent algorithm with fewer parameters and fast convergence. Based on the fact that the self-expression sub-problems are independent, a strategy was adopted to solve these sub-problems simultaneously on different processor cores, which brings the benefits of low computer resource consumption and fast running speed, it means that that the proposed algorithm is suitable for large scale clustering. Experiments on simulated data and Hopkins-155 motion segmentation dataset demonstrate that the proposed parallel SSC method on multi-core processors significantly improves the computational efficiency and ensures the accuracy when compared with ADMM.

The kernels of retest on existing railway are orbit line-shape segmentation and line-shape parameter optimization. Based on statistics of inertial angle measurement, an algorithm for line-shape segmentation of orbit plane and optimal line-shape parameter estimation was proposed. On the basis of change laws of orbit line-shape, the combined iterative method was utilized to calculate the optimal line-shape parameter of the orbit. The option of orbit plane line-shape was modeled as an optimization issue. Firstly, the orbit was roughly segmented by least square fitted slope change of fixed-curvature curves. Then, the orbit line-shape was fitted based on the measured data. Finally, the combined iterative method was applied to achieve precise segmentation and establish the optimal line-shape parameter. The simulation examples indicate that the proposed algorithm surpasses the existing artificial estimation algorithm which obtains the line-shape parameter fitting results based on two sets of different segmentation points, and has less differences from the results yielded by the exhaustion method. The Root Mean Square Error (RMSE) of the proposed algorithm is only higher than the exhaustion method by 4.93%, while the computational complexity is just 0.02% of that of the exhaustion method. The actual measurements on Xi'an metro line No.3 also have convinced the availability of the proposed algorithm.

The classic super-resolution algorithm via sparse coding has high computational cost during the reconstruction phase. In view of the disadvantages, a predictive sparse coding-based single image super-resolution method was proposed. In the training phase, the proposed method imposed a code prediction error term to the traditional sparse coding error function, and used an alternating minimization procedure to minimize the resultant objective function. In the testing phase, the reconstruction coefficient could be estimated by simply multiplying the low-dimensional image patch with the low-dimensional dictionary, without any need to solve sparse regression problems. The experimental results demonstrate that, compared with the classic single image super-resolution algorithm via sparse coding, the proposed method is able to significantly reduce the reconstruction time while maintaining super-resolution visual effect.

To solve the problem of low resource utilization in community health center, little contact between community health center and community residents, and difficulty for residents to participate in personal health management and medical care, an intelligent healthy community system was developed. With the increasing popular mobile devices, the system provided support for health record management, chronic disease management, immunization, appointment registration, medical information query and other services in community health center. It realized the data sharing and interaction among smart phones, tablet PCs and Hospital Information System (HIS), which allowed the residents to actively participate in personal health management. Now the system has been deployed in one community health center of Chengdu, it makes community residents convient to manage their personal health, and improves the work efficiency and service quality of community health center.

In three-dimensional sound reproduction with two speakers, Crosstalk Cancellation System (CCS) performance optimization often pay more attention to the effect independently by the factors such as inverse filter parameters design and loudspeaker configuration. A frequency-domain Least-Squares (LS) estimation approximation was proposed to use for the performance optimization. The relationship between these factors and their effect on CCS performance was evaluated systematically. To achieve the tradeoff of computing efficiency and system performance of crosstalk cancellation algorithm, this method obtained the optimization parameters. The effect of crosstalk cancellation was evaluated with Channel Separation (CS) and Performance Error (PE) index, and the simulation results indicate that these parameters can obtain good crosstalk cancellation effect.

With the increasing amount of data and expanding of application demand, the efficient organizational management and rapid processing speed of remote sensing data have become a bottleneck in the application of remote sensing technology. The earth partition theory and high performance computing provide a possible way to solve the above problem. Combined with global partition model, the conceptual model and data model of partition facet template were proposed based on partition facet of remote sensing image. A computing mode of partition facets based on templates was designed, and a small partition template database was established. A specific example of partition image data template applications was also given for validation. The experimental results demonstrate the feasibility of the data model and improve the efficiency of targets retrieval.

Bernstein’s Batch-factor algorithm can test B-smoothness of a lot of integers in a short time. But this method costs so much memory that it’s widely used in theory analyses but rarely used in practice. Based on splitting product of primes into pieces, a hierarchical batch-factor algorithm cloud framework was proposed to solve this problem. This hierarchical framework made the development clear and easy, and could be easily moved to other architectures; Cloud computing framework borrowed from MapReduce made use of services provided by cloud clients such as distribute memory, share memory and message to carry out mapping of splitting-primes batch factor algorithm, which solved the great cost of Bernstein’s method. Experiments show that, this framework is with good scalability and can be adapted to different sizes batch factor in which the scale of prime product varies from 1.5GB to 192GB, which enhances the usefulness of the algorithm significantly.

In order to remove the image blurring caused by reconstructing video frames independently frame by frame using traditional compressed video sensing method, this paper proposed a new approach to video compressed sensing based on motion estimation and motion compensation by combining the compressed sensing theory with related technology of MPEG standard video coding, so as to remove the spatial and temporal redundancy of video signal. This method took full account of the temporal correlations of video sequences and firstly compensated video frames using forward, backward and bidirectional prediction, then adopted the Backtracking-based Adaptive Orthogonal Matching Pursuit (BAOMP) algorithm to reconstruct the motion prediction residuals and finally reconstructed current frames. The experimental results indicate that the proposed method can gain a better video image quality compared with frame-by-frame reconstruction method and achieve a higher Peak Signal-to-Noise Ratio (PSNR).

With the development of computer technology and Internet, more and more office hosts have been connected to Internet, the threat of sensitive information leakage becomes serious. Therefore, it is extremely necessary to detect whether documents contain sensitive information. In order to solve the low precision and low recall problems caused by the traditional query expansion retrieval methods, this paper built an ontology of sensitive information for users interest, proposed a concept similarity query expansion algorithm based on the interest ontology, and described an experimental case to verify the feasibility of algorithm. The experimental results show that the proposed algorithm can improve the precision and recall of the traditional methods.

When building up the virtual battlefield system, to realize the three-dimensional (3D) detection range of radar in complex natural environment and complex environment of electronic interference, an improved support jamming model was proposed, according to the fundamental principle of Advanced Propagation Model (APM) and taking full consideration of the influence of electronic interference. This model mixed APM and electronic interference model together, and paid special attention to the refractive influence. Besides, this model could depict the double influence from complex natural and electronic interfering environments. Furthermore, a modified Marching Cube (MC) model, the triangles gained by MC being replaced by surface points and the interpolated points by middle points, was used to accelerate the process of visualization. According to the procedure of data gaining, data processing and data rendering, the 3D detection range of radar on specific electronic jamming environment was rendered via Visualization Toolkit (VTK).

A new data sharing scheme was proposed to solve the problem of error-tolerant search and fine-grained access control. This new scheme adopted the technology of locality-sensitive hashing and the predicate encryption, which allowed users to search for keywords in an error-tolerant manner, and modified the users access rights easily by updating the encrypted data. The computational complexity of updating is more optimized than the existing scheme. The theoretical analysis shows that the proposed solution is correct, safe and effective.

Pattern match search oriented to planar graphs is widely used in biological networks, social networks, fingerprint identification and image segmentation. Meanwhile, data extracted from those applications is inherently uncertain due to noise, incompleteness and inaccuracy, and query processing techniques of certain planar graphs cannot be applied to uncertain graph data. Therefore, in this paper, the pattern match query oriented to uncertain planar graphs was studied under the probabilistic semantics. Specifically, Uncertain Pattern Match (UPM) queries using the possible world semantics were investigated. Firstly, to avoid enumerating all possible worlds, a basic deterministic algorithm that can exactly compute UPM query was proposed. To further speed up the basic algorithm, an improved deterministic approach was developed based on tighten bounds. Secondly, a sampling algorithm that can quickly and accurately estimate matching probability was designed. The effectiveness of the proposed solutions for UPM queries was verified through extensive experiments on real uncertain planar graph datasets. Finally, UPM queries were applied to the segmentation on pulmonary CT images. The results show that the proposed approaches are superior to classical techniques of image segmentation.

To improve the storage efficiency of syntax, FLPM-B* tree, a full link pointer module B* tree, was designed. According to its structure characteristics, some algorithms such as module insert algorithm, reconstruction algorithm, and partition algorithm, were put forward. These algorithms make the FLPM-B* tree manipulated and efficient.