Focused on the issue that existing group recommendation methods take less account of the implicit estimation of socialization relationships among group members and the use of group consensus to reduce the influence of preference conflicts， a Group Recommendation method based on implicit Trust and group Consensus （GR-TC） was proposed. The method was divided into a recommendation phase and a consensus phase. In the recommendation phase， implicit trust values were mined based on preference information and social relationships among members. The members’ individual preferences and weights， and the initial group preferences were estimated. In the consensus phase， inconsistent members were identified by consensus measurement and identification rules， a maximum harmony optimization consensus model was built， and the group recommendation list was obtained by adjusting and updating the group preferences. Experimental results show that social relationships among members affect group recommendation results， reasonable selection of implicit trust weights improves the harmony of inconsistent members. Compared with the traditional consensus feedback mechanism， the implicit trust-induced maximum harmony consensus feedback mechanism has less adjustment cost and less impact on inconsistent members.

Unsupervised relation extraction aims to extract the semantic relations between entities from unlabeled natural language text. Currently， unsupervised relation extraction models based on Variational Auto-Encoder （VAE） architecture provide supervised signals to train model through reconstruction loss， which offers a new idea to complete unsupervised relation extraction tasks. Focusing on the issue that this kind of models cannot understand contextual information effectively and relies on dataset inductive biases， a Prompt-based learning based Unsupervised Relation Extraction （PURE） model was proposed， including a relation extraction module and a link prediction module. In the relation extraction module， a context-aware Prompt template function was designed to fuse the contextual information， and the unsupervised relation extraction task was converted into a mask prediction task， so as to make full use of the knowledge obtained during pre-training phase to extract relations. In the link prediction module， supervised signals were provided for the relation extraction module by predicting the missing entities in the triples to assist model training. Extensive experiments on two public real-world relation extraction datasets were carried out. The results show that PURE model can use contextual information effectively and does not rely on dataset inductive biases， and has the evaluation index B-cubed F1 improved by 3.3 percentage points on NYT dataset compared with the state-of-the-art VAE architecture-based model UREVA （Variational Autoencoder-based Unsupervised Relation Extraction model）.

In order to solve the problem of insufficient mining of potential association between remote nodes in human action recognition tasks， and the problem of high training cost caused by using multi-modal data， a multi-scale feature fusion human action recognition method under the condition of single mode was proposed. Firstly， the global feature correlation of the original skeleton diagram of human body was carried out， and the coarse-scale global features were used to capture the connections between the remote nodes. Secondly， the global feature correlation graph was divided locally to obtain the Complementary Subgraphs with Global Features （CSGFs）， the fine-scale features were used to establish the strong correlation， and the multi-scale feature complementarity was formed. Finally， the CSGFs were input into the spatial-temporal Graph Convolutional module for feature extraction， and the extracted results were aggregated to output the final classification results. Experimental results show that the accuracy of the proposed method on the authoritative action recognition dataset NTU RGB+D60 is 89.0% （X-sub） and 94.2% （X-view） respectively. On the challenging large-scale dataset NTU RGB+D120， the accuracy of the proposed method is 83.3% （X-sub） and 85.0% （X-setup） respectively， which is 1.4 and 0.9 percentage points higher than that of the ST-TR （Spatial-Temporal TRansformer） under single modal respectively， and 4.1 and 3.5 percentage points higher than that of the lightweight SGN （Semantics-Guided Network）. It can be seen that the proposed method can fully exploit the synergistic complementarity of multi-scale features， and effectively improve the recognition accuracy and training efficiency of the model under the condition of single modal.

To make use of the social relationship between experts and to consider the limited rationality of decision-making experts in group decision-making， a TODIM （TOmada de Decis?o Interativa Multicritério） group decision-making method under trust network was proposed. Firstly， according to the number of discussions of the experts， in each discussion， each expert would refer to his/her trustee’s decision matrix according to the degree of trust acceptance， and the decision matrices would be modified through information interaction and negotiation. Then， when the set number of expert discussions was met， the final group decision-making matrix was calculated. Finally， the TODIM group decision-making method under trust network and TODIM group decision-making method were applied to calculate the ranking results of different schemes. The ranking results were compared and analyzed， and the sensitivity analysis was performed on the number of expert discussions and trust acceptance. The case analysis results show that the TODIM group decision-making method under trust network can fully integrate trust network， ensure the multi-stage information interaction and feedback process in the decision-making process， and is superior to the general TODIM group decision-making method in comparison analysis and sensitivity analysis.

Refined short-term traffic flow prediction is the premise to ensure the rational decision making in Intelligent Transportation System （ITS）. In order to establish the lane-changing model of self-driving car， predict vehicle trajectories， and guide vehicle routes， the timely traffic flow prediction for each lane has become an urgent problem to solve. However， refined short-term traffic flow prediction faces the following challenges： first， with the increasing diversity of traffic flow data， the traditional prediction methods cannot meet the requirements of ITS for high precision and short time delay； second， training prediction model for each lane make a huge waste of resources. To solve the above problems， a refined short-term traffic flow prediction model combined Convolutional-Gated Recurrent Unit （Conv-GRU） with Grey Relational Analysis （GRA） was proposed to predict lane flow. Considering the characteristics of long training time and relatively short reasoning time of deep learning， a cloud-fog deployment scheme was designed. Meanwhile， to avoid training prediction models for each lane， a model migration deployment scheme was proposed， which only needs to train the prediction model of some lanes， and then the trained prediction models were migrated to the associated lane for prediction through GRA. Experimental results of extensive comparisons on a real-world dataset show that， compared with traditional deep learning prediction methods， the proposed model has more accurate prediction performance； compared with Convolutional-Long Short-Term Memory （Conv-LSTM） network， the model has shorter running time. Furthermore， the model migration is realized by the proposed model under the condition of ensuring high-precision prediction， which saves about 49% of training time compared to training prediction model for each lane.

Traditional hypernetwork model is biased towards the majority class, which leads to much higher accuracy on majority class than the minority when being tackled on imbalanced data classification problem. In this paper, a Boosting ensemble of cost-sensitive hypernetworks was proposed. Firstly, the cost-sensitive learning was introduced to hypernetwork model, to propose cost-sensitive hyperenetwork model. Meanwhile, to make the algorithm adapt to the cost of misclassification on positive class, cost-sensitive hypernetworks were integrated by Boosting. The proposed model revised the bias towards the majority class when traditional hypernetwork model was tackled on imbalanced data classification, and improved the classification accuracy on minority class. The experimental results show that the proposed scheme has advantages in imbalanced data classification.

The seizure detection is important for the localization and classification of epileptic seizures. In order to solve the problem brought by large amount of data and high feature space in EEG (Electroencephalograph) for quickly and accurately detecting the seizures, a method based on max-Relevance and Min-Redundancy (mRMR) criteria and Extreme Learning Machine (ELM) was proposed. The time-frequency measures by Short-Time Fourier Transform (STFT) were extracted as features, and the large set of features were selected based on max-relevance and min-redundancy criteria. The states were classified using the extreme learning machine, Support Vector Machine (SVM) and Back Propagation (BP) algorithm. The result shows that the performance of ELM is better than SVM and BP algorithms in terms of computation time and classification accuracy. The classification accuracy rate of interictal durations and seizures can reach more than 98%, and the computation efficiency is only 0.8s. This approach can detect epileptic seizures accurately in real-time.

Concerning the problem that Fireworks Explosion Optimization (FEO) algorithm is easy to be premature and has low solution precision, an elite Opposition-Based Learning (OBL) was proposed. In every iteration, OBL was executed by the current best individual to generate an opposition search populations in its dynamic search boundaries, thus the search space of the algorithm was guided to approximate the optimum space. This mechanism is helpful to improve the balance and exploring ability of the FEO. For keeping the diversity of population, the sudden jump probability of the individual to the current best individual was calculated, and based on it, the roulette mechanism was adopted to choose the individual which entered into the child population. The experimental simulation on five classical benchmark functions show that, compared with the related algorithm, the improved algorithm has higher convergence rate and accuracy for numerical optimization, and it is suitable to solve the high dimensional optimization problem.

Plane Wave Pseudopotential (PWP) Density Functional Theory (DFT) calculation is the most widely used method for material calculation. The projector calculation plays an important part in PWP-DFT calculation for the self-consistent iteration solution, while it often becomes a hinder to the speed-up of software. Therefore, according to the features of Graphic Processing Unit (GPU), a speed-up algorithm was proposed: 1) using a new parallel mechanism to solve the potential energy of nonlocal projector, 2) redesigning the distribution structure of data, 3) reducing the use of computer memory, 4) Proposing a solution to the related data problems of the algorithm. Eventually got 18-57 times acceleration, and reached the 12 seconds per step of the molecular dynamics simulation. In this paper, the testing time of running this model on GPU platform was analysed in detail, meanwhile the calculation bottleneck of the implementation of this method into GPU clusters was discussed

In order to improve the recommendation quality of recommendation system when the data are sparse, an improved collaborative filtering algorithm was proposed. Using a data mining algorithm, the sparse rating matrix was filled firstly. Afterwards user-similarities and their confidence factors were calculated using the complete filling matrix. Ultimately, the recommendation forecast was made. Comparative experiments on typical dataset show that the algorithm is able to achieve better results even with extremely sparse data.

Concerning the contradiction of accuracy and real-time in image fire detection, a fire image features selection and recognition algorithm based on rough set was proposed. Firstly, through in-depth study on the flame image features, the top edge of flame driven by the combustion energy is very irregular, and obvious vibration phenomenon occurres. But the lower edge is the opposite. Based on this feature, the upper and lower edges of the jitter projection ratio can be used as a flame from the edge shape regular interference. Then, the six striking flame features were chosen in order to create training samples. When fire classification ability was not affected, the feature classification table gained by experiment was used to reduce attributes of the training samples. And the reduced information systems attributes were applied to train a support vector machine model, and the fire detection was realized. Finally, this fire detection algorithm was compared to the traditional Support Vector Machine (SVM) fire detection algorithm. The results show that the presented algorithm reduces redundant attributes, eliminates the dimension of fire image features space, and decreases the data of training and testing in classifier in case rough set as a SVM classifier prefix system. While ensuring recognition accuracy, the algorithm improves fire detection speed.

In order to solve the problems of long handover delay and high packet loss rate existing in FMIPv6, an improved scheme named PI-FMIPv6 was designed. Information learning, proxy binding and the tunnel timer were introduced into it so as to complete the configuration of the New Care-of Address (NCoA), Duplicate Address Detection (DAD), Binding Update (BU) by advancing and managing the tunnel. π-calculus was used to define and deduce the mathematical model about PI-FMIPv6. It is proved that the optimized scheme PI-FMIPv6 is standard and precise. Furthermore, the simulation results from the NS-2 show that PI-FMIPv6 can reduce the handover delay by 60.7% and packet loss rate by 61.5% at least compared to FMIPv6, which verifies that the PI-FMIPv6 is superior to FMIPv6 and can better meet the real-time requirement.

Based on the statistical characteristics of satellite link error rate,this paper proposed a new link directly monitoring technology using variable length sequence. Using the analysis of bit error rate segmentation strategy and statistical principles of statistical confidence,the monitoring method determined the sequence of the link selection criteria through the training sequence length, the error simulation accuracy and reliability of monitoring and statistical analysis.Experimental results show that link directly monitoring technology using variable length sequence is effective to improve the range of link monitoring and reduce the computational complexity under a certain premise of the channel resources. It has certain advantages in the satellite to link monitoring.

Image-based fire detection can effectively solve the problems of large space fire detection contactlessly and rapidly. It is a new research direction in fire detection. Its essential issue is the classification of flames and disruptors. The ordinary detection methods are to extract one or a few characteristics of the flame in the image as a basis for identification. The disadvantages are to need a large number of experiential thresholds and the lower recognition rate by the inappropriate feature selection. Considering the entire characteristics of fire flame, a flame detection method based on Independent Component Analysis (ICA) and Support Vector Machine (SVM) was proposed. Firstly, a series of frames were pre-processed in RGB space. And suspected target areas were extracted depending on the flickering feature and fuzzy clustering analysis. Then the flame image features were described with ICA. Finally, SVM model was used in order to achieve flame recognition. The experimental result shows that the proposed method improves the accuracy and speed of image fire detection in a variety of fire detection environments.

To solve the multi-spot inventory sharing problem in an emergency system, emergency transportation strategy was studied in a system with random fuzzy demand in this paper through a multi-product and three-echelon emergency supply system. When the stockout happened, the nearest emergency lateral transshipment principle and partial inventory sharing strategy among the spots were permitted to satisfy the demand, and the model for the total cost expectation of random fuzzy demand was developed according to it, taking account of the service time constraints and the spots' storage space limitation. An advanced computing method combining Particle Swarm Optimization (PSO) and Simulated Annealing (SA) algorithm, called PSO-SA algorithm, was proposed to calculate the model, and the effects on the partial transshipment with the variation of the transshipment trigger inventory level, the per-item transshipment time and the inventory storage space were analyzed through a numerical example. The availability of the proposed algorithm and the model applicability were verified at last.

Most conventional software can not implement communication protection by modifying compiled binary code. Concerning this problem, this paper proposed a general LSP-based confidential communication model. The model introduces the optimization of the linear Hash table to store communication information, and realizes communication protection function on LSP layer by converted Winsock network model and hooked system function. The experimental results show that the proposed model can efficiently realize confidential communication between client and server.

The shortened LDPC codes based on the current IEEE802.16e Standard, has plenty of small girth. In order to resolve this problem, this paper presents a new scheme of designing the shortened LDPC codes. In the proposed scheme, the design has modified the sub-parity check matrix under the frame of the IEEE802.16e Standard. The proposed check matrix with the spreading factor (zf=48) is constructed with quasi-cyclic matrix method and finite geometry method. Moreover, parity check points was searched the node degree in the way of synchronous sequential, the proposed sub- parity check matrix has minor girth-6 and no girth-4. The results of simulation in the AWGN channels show that the improved short code with the code rate approaching to 0.5, can fast encode as excellent as the IEEE802.16e protocol contains，moreover, a good BER performance of the proposed code is 1.1dB more than the Shannon limit of this channel.

The paper proposed an accelerating way of face detection based on dual-threshold cascade classifiers. First, it applied Gabor filter to extract the face-like features that were retained by template matching, then put eigenvectors extracted by the way of Principal Component Analysis (PCA) into the BP neural network as first classifier, then used dual-threshold to decide face or non-face on output end, and put the face or non-face of midway between up and down threshold into the AdaBoost classifier as the second classifier to decide. In this way, it can improve the detection rate and reduce the false rate while speeding up the detection speed. The experimental results prove that the precision of cascade classifier of face detection based on dual-threshold is superior to the classifier of single threshold.

A distributed simulation approach was proposed for discrete-events simulation with considerable amounts of events between logical processes. The proposed approach employed a time-driven method to simulate occurrence of discrete-events, using Remote Procedure Call (RPC) to describe the interaction between simulation members. In this approach, barrier synchronization objects were deployed for time synchronization in simulation advancement, in order to ensure the correctness of the causal ordering. Results obtained from the experiments show that the proposed approach can correctly and promptly handle large number of events, providing accuracy guarantee and efficiency improvement of the simulation model.

Nowadays, as low carbon economy has been advocated worldwide, the power consumption of embedded software has become a critical factor in embedded system design. The hardware/software partitioning is an important method of embedded software power optimization. Firstly, this paper constructed a hardware/software bi-partitioning model with the goal of embedded software power consumption under the constraints of performance; then, a hybrid algorithm was proposed based on the fusion of discrete Hopfield Neural Network (HNN) and Tabu Search (TS), in which HNN as the main method could quickly obtain a feasible solution of partitioning, and the TS algorithm could "taboo" the current solution and transferred to the other minimum points that could jump out from the local optimal solution. Lastly, the experimental results show that the proposed algorithm posses better time performance and higher probability of acquiring the global optimal solution in contrast with other similar algorithms.

A new improved algorithm for image denoising was proposed by analyzing the Total Variational (TV) model. According to the viewpoint of Maximum A Posteriori (MAP) and Markov Random Field (MRF) theory, a generalized variational functional model was deduced. And the Lagrange multiplier λ used for balancing the data fidelity term and regularized term was adaptively improved. An edge preserving potential function was adopted, which had good robustness to noises; finally an iterative algorithm was exploited to solve the energy functional combining weighted gradient descent flow and semi-point scheme. Experimental results show that the proposed model has good performance in image denoising. It is obviously superior to the conventional variational model in both visual effect and PSNR.

An integrated algorithm based on DV-Hop was designed. A Location Correction Vector（LCV）was constructed by the differences between estimated distances and range measurements. Nodes were clustered when the anchors were the heads of clusters, object function expressing total distances error was constructed in a cluster, Particle Swarm Optimization (PSO) was used to solve the minimization problem, and then correction steps of all member nodes were obtained. The value of location correction equaled the product of LCV and steps, then extra location correction had been executed by using the relative positions among edge nodes of neighbor clusters. Simulation results show that the localization error of DV-Hop is reduced by 75%, and it is also applicable to low-density networks.

To efficiently simplify the densely sampled point model, a point sample data reduction method was proposed based on the mean-shift clustering algorithm. Local mode centroids were calculated by mean-shift iterative process. These mode centroids substituting for ambient data points were used to simplify the model. Experiment results show that the algorithm can effectively simplify the densely sampled point model, the reduction speed is fast, and the simplified model can preserve the original geometric shape.

A new algorithm based on continuous mutation was proposed. The mixed selection was used for choosing individuals. Through the combination of the phased crossover and the cosine adaptive crossover, the double adaptive crossover got the crossover probability. The continuous mutation strategy used continuous process from crude search to precise search. Numerical experiments show that the new algorithm is more effective in realizing the high convergence speed, convergence precision, reducing the convergence algebra and good at keeping the stability of the adaptive genetic algorithm.

Defects in welding seam X-ray images include similar circular defects and similar bar defects. According to the theory of multiscale edges detection, the conclusion was drawn that in appropriate scale, circular defects are considered as roof edges in two directions that are orthogonal, while bar defects are viewed as roof edges in one direction. Wavelet scales can be confirmed by region homogeneity measure in this paper. LOG operator and filter scales of steerable filters were actively ascertained in the defect areas so that the two types of defects could be segmented. Theoretical analysis and experimental results show that this algorithm is effective.

A revision method was proposed for the offset of separation hyperplane of binary-classification imbalanced data in Support Vector Machine (SVM). Firstly, the principal values were found respectively of the two classes of samples in feature space by using Kernel Principal Component Analysis (KPCA). Secondly, one penalty proportion was given based on the information provided by the sizes of the two sample data and their values. Finally, a new separation hyperplane was generated through the optimization training. The hyperplane revised the error of the standard support vector machines. Experiment results prove the validity of the method. Compared with standard support vector machines, the proposed method can not only balance but also decrease the classification error.