During the evaluation and review process of scientific and technological projects， the quality of evaluation and review conducted by experts significantly influences the accuracy and credibility of the final evaluation results. To ensure the fairness and objectivity of the evaluation results， it is necessary to conduct a counter-evaluation of the evaluation experts. By analyzing three aspects including the basic personal situation， professional level and evaluation performance of evaluation experts， an evaluation index system for counter-evaluation of experts was constructed. Based on this， an expert counter-evaluation model that combined three-way decisions and entropy weight based Technique for Order Preference by Similarity to Ideal Solution （TOPSIS） was proposed， which used the three-way decision theory to solve the problem of weight distortion due to the over-reliance of entropy weight method on data. When the index weights were abnormal， the group of experts to be evaluated was divided into positive， negative and boundary domains according to the threshold， and then the expert-in， expert-out and delayed evaluation strategies were adopted respectively. Once the index weights were normal or corrected， TOPSIS was used to rank the evaluation experts. Through empirical analysis of the historical evaluation and review data of scientific and technological projects from an enterprise， it can be seen that the proposed model can integrate the empirical judgment of decision makers with inherent information of the experts to be evaluated to realize the unity of subjectivity and objectivity， ensuring a scientific and fair evaluation of the evaluation experts， and providing decision-making references for construction of a high-quality expert database.

In order to solve the occlusion problem of student expression recognition in complex classroom scenes， and give full play to the advantages of deep learning in the application of intelligent teaching evaluation，a student expression recognition model and an intelligent teaching evaluation algorithm based on deep attention network in classroom teaching videos were proposed. A video library， an expression library and a behavior library for classroom teaching were constructed， then， multi-channel facial images were generated by cropping and occlusion strategies. A multi-channel deep attention network was built and self-attention mechanism was used to assign different weights to multiple channel networks. The weight distribution of each channel was restricted by a constrained loss function， then the global feature of the facial image was expressed as the quotient of the sum of the product of the feature times its attention weight of each channel divided by the sum of the attention weights of all channels. Based on the learned global facial feature， the student expressions in classroom were classified， and the student facial expression recognition under occlusion was realized. An intelligent teaching evaluation algorithm that integrates the student facial expressions and behavior states in classroom was proposed， which realized the recognition of student facial expressions and intelligent teaching evaluation in classroom teaching videos. By making experimental comparison and analysis on the public dataset FERplus and self-built classroom teaching video datasets， it is verified that the student facial expressions recognition model in classroom teaching videos achieves high accuracy of 87.34%， and the intelligent teaching evaluation algorithm that integrates the student facial expressions and behavior states in classroom achieves excellent performance on the classroom teaching video dataset.

For low server utilization and serious energy consumption waste problems in cloud computing environment, an energy-efficient strategy for dynamic management of cloud storage replica based on user visiting characteristic was put forward. Through transforming the study of the user visiting characteristics into calculating the visiting temperature of Block, DataNode actively applied for sleeping so as to achieve the goal of energy saving according to the global visiting temperature.The dormant application and dormancy verifying algorithm was given in detail, and the strategy concerning how to deal with the visit during DataNode dormancy was described explicitly. The experimental results show that after adopting this strategy, 29%-42% DataNode can sleep, energy consumption reduces by 31%, and server response time is well. The performance analysis show that the proposed strategy can effectively reduce the energy consumption while guaranteeing the data availability.

Given the improper threshold, Sobel operator could easily lead to the loss of image edges or produce broad pseudo-edges. To solve these problems, appropriate threshold was selected by using the constraint of variance firstly, and then mathematical morphology thinning algorithm was utilized to thin the edge images detected by Sobel operator. The experimental results show that this method can bring about satisfactory thinning effects while retaining the original edge information.

Due to the uncertainty of complex networks, traditional community structures division algorithm of the complex network could easily lead to premature convergence and decreased accuracy. And because of the large amount of computation, time complexity is high. To overcome the above shortcomings, the paper adopted GEP's global search ability and adaptability, and other characteristics with parallel calculations, optimized the network structure of the division of community, and proposed a community structure division algorithm of complex network based on GEP, and verified the validity of the new algorithm by experiment. The new algorithm has more accurate community division of the complex network in the case of no prior information.

Handover procedure of modern cellular wireless networks depends on IP-based technology. IETF Proxy Mobile IPv6 (PMIPv6) protocol guarantees the Quality of Service (Qos) in fast handover moving while it does not support realtime communications between two mobile nodes. Therefore Cluster Mobile Node (CMN) algorithm was proposed to reduce handover delay in the system by applied Media Independent Handover (MIH) Technology. Also, the algorithm extended PMIPv6 protocol with an Aggregated Proxy Binding Update (A-PBU) scheme in the paper. Finally, the network model and mobile model were simulated and the effectiveness of handover delay was analyzed. Quantitative results show a significant reduction in handover delay compared with the original handover algorithm.

Increasing diversity and complexity of the computing environments result in various security requirements. MLS security policy only aims at confidentiality assurance, in less consideration of integrity assurance and weakness in channel control. To handle that the trusted subjects have many security shortcomings of MLS model, a multipolicy views security model (MPVSM) was presented. Based on the MLS model, MPVSM combined the domain and type attributes to the model, to enforce the channel control policy, make permission management more finegrained and enhance the ability to confine the permission of the trusted subjects. MPVSM was also able to enforce multipolicy views in operating system in a flexible way. The implementation of the MPVSM model in our prototype trusted operating system Nutos was also introduced.