Focusing on the challenge task for mining complementary information in different levels of features in the deep subspace clustering problem， based on the deep autoencoder， by exploring complementary information between the low-level and high-level features obtained by the encoder， a Diversity Represented Deep Subspace Clustering （DRDSC） algorithm was proposed. Firstly， based on Hilbert-Schmidt Independence Criterion （HSIC）， a diversity representation measurement model was established for different levels of features. Secondly， a feature diversity representation module was introduced into the deep autoencoder network structure， which explored image features beneficial to enhance the clustering effect. Furthermore， the form of loss function was updated to effectively fuse the underlying subspaces of multi-level representation. Finally， several experiments were conducted on commonly used clustering datasets. Experimental results show that on the datasets Extended Yale B， ORL， COIL20 and Umist， the clustering error rates of DRDSC reach 1.23%， 10.50%， 1.74% and 17.71%， respectively， which are reduced by 10.41， 16.75， 13.12 and 12.92 percentage points， respectively compared with those of Efficient Dense Subspace Clustering （EDSC）， and are reduced by 1.44， 3.50， 3.68 and 9.17 percentage points， respectively compared with Deep Subspace Clustering （DSC）， which indicates that the proposed DRDSC algorithm has better clustering effect.

Aiming at the problems of low detection accuracy and poor generalization ability of Wreless Sensor Network （WSN） intrusion detection methods on imbalanced datasets with discrete high-dimensional features， an intrusion detection method for WSN based on Bidirectional Circulation Generative Adversarial Network was proposed， namely BiCirGAN. Firstly， Adversarially Learned Anomaly Detection （ALAD） was introduced to improve the understandability of the original features by reasonably representing the high-dimensional， discrete original features through the latent space. Secondly， the bidirectional circulation adversarial structure was adopted to ensure the consistency of bidirectional circulation in real space and latent space， thereby ensuring the stability of Generative Adversarial Network （GAN） training and improving performance of anomaly detection. At the same time， Wasserstein distance and spectral normalization optimization methods were introduced to improve the objective function of GAN to further solve the problems of mode collapse of GAN and lack of diversity of generators. Finally， because the statistical properties of intrusion attack data changed in an unpredictable way over time， a full connection layer network with Dropout operation was established to optimize the anomaly detection results. Experimental results on KDD99， UNSW-NB15 and WSN_DS datasets show that compared to Anomaly detection with GAN （AnoGAN）， Bidirectional GAN （BiGAN）， Multivariate Anomaly Detection with GAN （MAD-GAN） and ALAD methods， BiCirGAN has a 3.9% to 33.0% improvement in detection accuracy， and the average inference speed is 4.67 times faster than that of ALAD method.

Existing certificateless proxy signature schemes based on classical number theory problem assumptions cannot resist to quantum computer attacks， and when these schemes are applied to systems with a large number of users， there are limitations such as single point of failure and low scalability. Aiming at these problems， a lattice-based hierarchical certificateless proxy signature scheme was proposed. Firstly， the rejection sampling technology and trapdoor-free technology were used to improve the computational efficiency of key generation. Secondly， the mutual authentication was performed by the original signers and proxy signers at different levels by exchanging randomly selected matrices， and then the proxy authorization was realized. Finally， the security of this scheme was proved under the of the Small Integer Solution （SIS） hard problem assumption in the random oracle model. Compared with the existing proxy signature schemes， the proposed scheme allows signers coming from different levels and belonging to different Key Generation Centers （KGCs）. The performance evaluation experimental results show that in the proposed scheme， the public key size is a constant， the overhead of proxy signature and verification is independent of the level， and the proxy key size and the signature size are not hierarchical linear quantities， so that this scheme can better meet the needs of large-scale distributed heterogeneous networks for load balancing， and is efficient and feasible.

In order to accelerate rendering of 3D scenes by path tracing， a visual?saliency?driven reuse algorithm of indirect lighting in 3D scene rendering was proposed. Firstly， according to the characteristics of visual perception that the regions of interest have high saliency， while other regions have low saliency， a 2D saliency map of the scene image was obtained， which consists of color information， edge information， depth information and motion information of the image. Then， the indirect lighting in the high?saliency area was re?rendered， while the indirect lighting of the previous frame was reused in the low?saliency area under certain conditions， thereby accelerating the rendering. Experimental results show that the global lighting effect of the image generated by this method is real， and the rendering speed of the method is improved in several experimental scenes， and the speed can reach up to 5.89 times of that of the high?quality rendering.

For the intelligent dialogue system of customer service robots in power supply business halls, a large-scale dataset of power business user intents was constructed. The dataset includes 9 577 user queries and their labeling categories. First, the real voice data collected from the power supply business halls were cleaned, processed and filtered. In order to enable the data to drive the study of deep learning models related to intent classification, the data were labeled and augmented with high quality by the professionals according to the background knowledge of power business. In the labeling process, 35 types of service category labels were defined according to power business. In order to test the practicability and effectiveness of the proposed dataset, several classical models of intent classification were used for experiments, and the obtained intent classification models were put in the dialogue system. The classical Text classification model-Recurrent Convolutional Neural Network (Text-RCNN) was able to achieve 87.1% accuracy on this dataset. Experimental results show that the proposed dataset can effectively drive the research on power business related dialogue systems and improve user satisfaction.

Aiming at the problems that simple Recurrent Neural Network (RNN) cannot memorize information for a long time and single Convolutional Neural Network (CNN) lacks the ability to capture the semantics of text context, in order to improve the accuracy of text classification, a sentiment analysis model G-Caps (Gated Recurrent Unit-Capsule) was proposed, which combines Gated Recurrent Unit (GRU) and capsule features. First, the contextual global features of the text were captured through GRU in order to obtain the global scalar information. Second, the captured information was iterated through the dynamic routing algorithm at the initial capsule layer to obtain the vectorized feature information representing the overall attributes of the text. Finally, the features were combined in the main capsule part to obtain more accurate text attributes, and the sentiment polarity of the text was analyzed according to the intensity of each feature. Experimental results on the benchmark dataset MR (Movie Reviews) showed that compared with the CNN + INI (Convolutional Neural Network + Initializing convolutional filters) and CL_CNN (Critic Learning_Convolutional Neural Network) methods, G-Caps had the classification accuracy increased by 3.1 percentage points and 0.5 percentage points respectively. It can be seen that the G-Caps model effectively improves the accuracy of text sentiment analysis in practice.

Aiming at the problem of slowness of the current meta-heuristic algorithms when solving Travelling Salesman Problem (TSP) in combinatorial optimization problems, a multi-scale adaptive quantum free particle optimization algorithm was proposed based on the inspiration of the wave function in quantum theory. Firstly, the particles representing the city sequences were randomly initialized in the feasible region as the initial search centers. Then, the new solution was obtained by taking each particle as the center to perform the sampling with uniformly distributed function and exchanging the city numbers in the sampling positions. Finally, according to the comparison result of the new solution with the optimal solution in the previous iteration, the search scale was adaptively adjusted, and the iterative search was carried out at different scales until the end condition of the algorithm was satisfied.The algorithm was compared with Hybrid Particle Swarm Optimization (HPSO) algorithm, Simulated Annealing (SA), Genetic Algorithm (GA) and Ant Colony Optimization(ACO) algorithm on TSP. The experimental results show that the multi-scale quantum free particle optimization algorithm is suitable for solving combinatorial optimization problems, and increases the solving speed by over 50% on average compared with the current better algorithms on the TSP datasets.

In order to solve the intelligence extraction of information in the era of remote sensing big data,it is important to build the model and method of intelligent information analysis fitting the intrinsic characteristics of remote sensing data. To meet the demand of universal remote sensing intelligent acquisition of large-scale water information,an intelligent extraction method of remote sensing water information based on visual selective attention mechanism and AdaBoost algorithm was proposed. Firstly,by the optimization design of RGB color scheme of remote sensing multi-feature index,the enhancement and visual representation of the water information image features were realized. Then,in HSV color space,the key node information of the chromatic aberration distance image was used to construct the classification feature set,and AdaBoost algorithm was used to construct the water recognition classifier. On this basis,the category that the water belongs to was automatically recognized from the image color clustering result,so as to realize the intelligent extraction of water information. Experimental results show that the proposed method has the water information extraction results improved on Leak Rate(LR)and Composite Classification Accuracy(CCA). At the same time,the proposed method not only effectively reduces the dependence on high quality training samples,but also has good performance on the recognition of temporary water areas such as water with high sediment concentration at wet season and submerged area caused by flooding.

In order to evaluate the reliability and fault-tolerant ability of multi-processor system which takes hypercubes as underlying networks, combining the fact that structural faults often occur when the system is invaded by computer viruses, three-length-path structure connectivity and substructure connectivity of the n-cube network were investigated. Firstly, by using the three-length-path structure-cut of the n-cube network, an upper bound of three-length-path structure connectivity of the network was obtained. Secondly, by using an equivalent transformation or a reductive transformation of the three-length-path substructure-set of the n-cube network, a lower bound of three-length-path substructure connectivity of the network was obtained. Finally, combining with the property that three-length-path structure connectivity of a network is not less than its three-length-path substructure connectivity, it was proved that both three-length-path structure connectivity and substructure connectivity of a n-cube network were half of n. The results show that to destroy the enemy's multi-processor system which take the n-cubes as underlying networks under three-length-path structure fault model, at least half of n three-length-path structures or substructures of the system should be attacked.

Although the click farming on e-commerce platform improves the store profits to some extent, but it raises the promotion cost of e-commerce platform, which leads to a serious problem of reputation security, and on the other hand, it misleads consumers with property loss. To solve these problems, an intelligent method named SVM-NB was proposed for detecting the click farming on e-commerce platform for users, and a method of constructing characteristics of click farming was also put forward. Firstly, the relevant data of commodity were collected to create an eigenvalue database. Then a classifier was established based on Support Vector Machine (SVM) algorithm with supervised learning, so as to judge the result of click farming. Finally, the click farming probability of goods was calculated by using Naive Bayes (NB), which can provides users with a reference for their shopping. The reasonality and accuracy of the proposed SVM-NB method was validated by K-fold cross validation algorithm, and the accuracy reached 95.0536%.

Error Back Propagation (BP) algorithm is iterative. How to implement it using iterative MapReduce framework was studied. To avoid the shortage of the traditional MapReduce framework that task must submit repeatedly in iterative program, a transmitting module was added into the iterative MapReduce framework. The training samples obtained from the simulation of the control system in the K/TGR146 radio switch were trained on Hadoop using the traditional framework and the iterative framework respectively. The training of BP algorithm based on the iterative framework is more than 10 times faster than BP algorithm based on the traditional framework, and its accuracy raises by 10%-13%. The iterative framework can effectively reduce the training time and avoid submitting task repeatedly in iterative calculation.

In view of the disadvantages of small link connectivity, restricted edge connectivity and weak fault tolerance, a kind of augmented bubble-sort network was designed by adding some links to the original bubble-sort network. By constructing a minimum link cut of the n dimensional augmented bubble-sort network, the link connectivity of the n dimensional augmented bubble-sort network was proved to be n, which implied that any two nodes are still connected even deleting n-1 links. The restricted link connectivity of the n dimensional augmented bubble-sort network was proved to be 2 n-2. Therefore, any two nodes of the n dimensional augmented bubble-sort network were still connected even deleting 2 n-3 links if the removal of these links doesn't result in singletons. Based on above results, the example rusults show that the augmented bubble-sort networks is better than bubble-sort network.

Since most malwares are designed using decentralized architecture to resist detection and countering, in order to fast and accurately detect Peer-to-Peer (P2P) bots at the stealthy stage and minimize their destructiveness, a real-time detection system for stealthy P2P bots based on statistical features was proposed. Firstly, all the P2P hosts inside a monitored network were detected using means of machine learning algorithm based on three P2P statistical features. Secondly, P2P bots were discriminated based on two P2P bots statistical features. The experimental results show that the proposed system is able to detect stealthy P2P bots with an accuracy of 99.7% and a false alarm rate below 0.3% within 5 minutes. Compared to the existing detection methods, this system requires less statistical characteristics and smaller time window, and has the ability of real-time detection.

In we media platform such as microblog, emergency has such characteristics as suddenness and having multiple bursting points. Thus, it brings difficulty to emergency detection. Thus, this paper proposed a method of bursty events detection based on sentiment filter. Firstly, the topic was mapped as a hierarchical model according to the method. Then, dynamic adjustment of the model characteristics was made in a timing-driven way so as to detect the new topics of the information. Based on it, the method analyzed the user's emotional attitude toward such topics. The topics were divided into positive and negative emotion tendencies according to the user's emotional attitude. Additionally, the topic full of negative emotion tendency was regarded as emergent topic. The experimental results show that the accuracy and recall of the proposed method are all increased about 10% compared with baseline.

For the difficulty of expressing spatial context in classification of high resolution remote sensing imagery, a new multi-scale Conditional Random Field (CRF)model was proposed here. Specifically, a given image was represented as three superpixel layers respectively being region, object and scene from fine to coarse firstly. Then features were extracted layer-by-layer, and those features from the three layers were associated with each other to form a feature vector for each node in region layer. Secondly, Support Vector Machine (SVM) was adopted to define association potential function, and Potts model weighted by feature contrast function was used to define interaction potential function of CRF model, thus a layer-by-layer feature associative and multi-scale SVM-CRF model was formed. To confirm the effectiveness of the proposed model in classification, experiments on two complex scenes from Quickbird remote sensing imagery were developed. The results show that the proposed model achieves an improved accuracy averagely 2.68%, 2.37%, 3.75% higher than that of SVM-CRF model based on either region, object or scene layer, also it consumes less time in classification.

It is difficult for the existing methods to get overall sentiment orientation of the comment text. To solve this problem, the method of multi-document sentiment summarization based on Latent Dirichlet Allocation (LDA) model was proposed. In this method, all the subjective sentences were extracted by sentiment analysis and described by LDA model, then a summary was generated based on the weight of sentences which combined the importance of words and the characteristics of sentences. The experimental results show that this method can effectively identify key sentiment sentences, and achieve good results in precision, recall and F-measure.

In order to solve the problem of insufficient training corpus for extracting attribute relation from Chinese encyclopedia, a weakly supervised method was proposed, which needed minimal human intervention. First, semi-structured attribute relations from Chinese encyclopedia entry infoboxes were used to tag entry texts for obtaining training corpus. Second, the optimized training corpus was obtained based on Naive Bayesian theory. Third, Conditional Random Field (CRF) was used to form attribute relation extraction model. The evaluation of F-score on the Hudong encyclopedia datasets was 80.9%. The experimental result shows that this method can enhance the quality of training corpus and runs a better extraction performance.

The thorough research on ant colony algorithm points out that the ant colony algorithm has superiority in solving large nonlinear optimization problem. Through careful analysis of the deficiencies that genetic algorithm and particle swarm algorithm solve the problem of vehicle dispatching system, based on the advantage of ant colony algorithm and the own characteristics of vehicle dispatching system, the basic ant colony algorithm was improved in the paper, and the algorithm framework was created. Based on the linear programming theory, the article established mathematical model and operation objectives and constraints for vehicle dispatching system, and got the optimal solution of vehicle dispatching system problem with the improved ant colony algorithm. According to the optimal solution and the dispatching criterion real-time scheduling was achieved. The article used Java language to write a simulation program for comparing the improved particle swarm optimization algorithm and ant colony algorithm. Through the comparison, it is found a result that the improved ant colony algorithm is correct and effective to solve the vehicle dispatching optimization problem.

In order to measure the fault tolerance ability of the parallel computers which take the k-ary n-cube as underlying topology, by constructing the minimum node set whose removal will lead to every k-ary 1-cube in the k-ary n-cube faulty, a recursive algorithm for finding the k-ary 1-cube subnet preclusion node cut of the k-ary n-cube was proposed. It is proved that at least kn-1 nodes need to be damaged if a rival wants to destroy all k-ary 1-cubes in the k-ary n-cube. The result indicates that there are still undamaged k-ary 1-cubes in the parallel computers which take the k-ary n-cube as underlying topology if the number of the faulty nodes does not exceed kn-1-1.

Concerning the high computation complexity and low efficiency in traditional stereo matching method based on belief propagation, a fast-converging algorithm was proposed. When calculating the confidence level of each pixel, the algorithm only utilized the information translated from the neighboring pixels in an adaptive support window, while ignoring the impact of the pixels beyond the window. The experimental results show that the proposed algorithm can reduce 40% to 50% of computation time while maintaining the matching accuracy. Therefore, it can meet the real-time requirement for stereo matching.

Network management becomes more difficult for the increase of data transmission speed and network complexity, so the paper presented an intelligent multi-Agent model for virtual resource, described the process of the multi-Agent to virtual resource, and discussed the processing mechanism of different Agent. The proposed model was able to analyze social media resources in real-time by user context and system state. It automatically allocated resources suitable for users according to the virtual resource usage type and the information demand analysis of user context. The model was evaluated by dynamic scheduling method of virtual resources in cloud computing and the MovieLens system. The results show that the proposed model has better performance, can achieve the dynamic scheduling and load balancing of virtual resource, so that users can utilize efficiently virtual resource in the cloud computing.

The existing Web services discovery method has limitations in time cost and accuracy because it does not make full use of the user context. Firstly, the clustering of similar user context was implemented to greatly reduce retrieval range of Web services. Secondly, based on this, making use of the current users' preference information and the history users' QoS-aware data, a method of Web services discovery based on history users' QoS-awareness was proposed. Finally, the comparison to other Web services methods indicates that this method is better than several other methods both in time cost and accuracy of Web services discovery.

This paper proposed an iterative carrier phase and channel estimation algorithm for 16APSK in Bit-Interleaved Coded Modulation with Iterative Decoding (BICM-ID). The algorithm was based on Maximum Likelihood （ML） estimation, which made use of the decision information provided by the decoder and exchanged information between phase, channel estimation and decoder through iteration, so as to combine phase, channel estimation with decoder. The simulation results show the algorithm can get the performance only 0.5dB lower than the ideal one, and can effectively estimate the phase difference from -20 degree to 20 degree.

Many different communication radius of the communication nodes that may cause an unstable network performance can be easily found in Epidemic Routing (ER) network. A network model that combines network coding and epidemic routing can solve this problem. Compared with the traditional epidemic routing, the Network Coding Based Epidemic Routing (NCER) can transmit packets with network coding. In order to compare the performances of the ER and NCER, a probability model of the transmission delay of the network was built. The comparative results between the two protocols with the probability model above show that NCER can be more efficient and stable than ER. The correctness of this probability model has been proved in the simulation. Finally, according to the model evaluation results, a scheme has been given to reduce the network transmission delay.

In order to improve spectrum usage in Cognitive Radio System (CRS), and reduce channel switching frequency, a new prediction mechanism was designed, which was used chaotic neural network to analyze and predict the last time of channel status. Simulation results show that the prediction accuracy can reach 90%, thus the effectivess of this new prediction mechanism was proved.

In order to compute approximate geodesic efficiently between two points on point cloud, a weighted graph was constructed by splitting point cloud along the Cartesian coordinate axes, thus initial approximate geodesic between any two given points could be computed out using Dijkstra's algorithm. Then the conjugate gradient method was adopted to minimize the energy function defined; finally, approximate geodesic could be obtained after some iterative steps. This proposed algorithm avoids meshing or reconstructing the point cloud to be local or global surface, and it is suitable for computing geodesic on large scale point cloud.

In order to discover the special behaviors of Sub Complex System (SCS) under intervention, the authors proposed the concept of contrast inequality, proposed and implemented the algorithm for mining the segmentwise intervention; by imposing perturbance intervention on SCS, the authors proposed and implemented the causality discovery algorithm. The experiments on the real data show that segmentwise intervention algorithm discovers new intervention rules, and the causality discovery algorithm discovers the causality relations in the air pollution data set, and both are difficultly discovered by traditional methods.

Arnold cat transformation is widely applied and has the best scrambling effect. However, it has some disadvantages, such as small key quantities and poor generalizability. From the scrambling essential, the concept of even scrambling was presented. At the same time, an algorithm of image position scrambling on the basis of improving Arnold cat transformation was put forward. The experimental results show that the key quantities and scrambling effect of the proposed algorithm are improved obviously, and it can resist cut attack.

As the civil aviation develops, organization rebuilding and task reassigning will be repeating in a long period. Workflow technology provides separation of the business procedure logic and its IT operational support, enabling subsequent changes to be incorporated into the procedural rules defining the business process. Therefore, workflow technology is applied in civil airport ELV (Electronic Low Voltage) information system based on middleware service layer, to avoid unnecessary overlapping development of computer management and control system. Firstly, the construction of ELV system is was analyzed and content of workflow modeling is was introduced. Secondly, three kinds of ELV system architecture based on workflow technology are were discussed. Finally, the realization of distributed prototype system taking on main control workflow engine is was described.

Bayesian network is a forceful tool to practise inference of uncertainty. It combines graphic theories and probability ones, which can conveniently express and calculate the probability of interesting events and at the same time provide a compact, visual and effective graphic expression for the dependant relationship among the entities. On the basis of testing information independence theory, the test of CI（conditional independence） was carried out on all the joints in the Bayesian network to find out the conditionally dependant relations among them. Then an effective algorithm of Bayesian network structural learning was worked out, which only needed CI testing of O（N ²） times.