In order to solve the problem of insufficient available training data in the classification task of breast mass and calcification, a multi-view model based on secondary transfer learning was proposed combining with imaging characteristics of mammogram. Firstly, CBIS-DDSM (Curated Breast Imaging Subset of Digital Database for Screening Mammography) was used to construct the breast local tissue section dataset for the pre-training of the backbone network, and the domain adaptation learning of the backbone network was completed, so the backbone network had the essential ability of capturing pathological features. Then, the backbone network was secondarily transferred to the multi-view model and was fine-tuned based on the dataset of Mianyang Central Hospital. At the same time, the number of positive samples in the training was increased by CBIS-DDSM to improve the generalization ability of the network. The experimental results show that the domain adaption learning and data augmentation strategy improves the performance criteria by 17% averagely and achieves 94% and 90% AUC (Area Under Curve) values for mass and calcification respectively.

Aiming at the concept drift problem, a classification learning model with the characteristics of data changing progressively over time was constructed, and a Gradual Multiple Kenerl Learning method (G-MKL) based on Gradual Support Vector Machine (G-SVM) was proposed. In this method, with Support Vector Machine (SVM) used as the basic classifier, multi-interval sub-classifier coupling training was carried out and the incremental method of constraining sub-classifier was used to adapt the model to the gradual change of data. Finally, multiple kernels were integrated into SVM solution framework in a linear combination manner. This method integrated the advantages of different kernel functions and greatly improved the adaptability and validity of the model. Finally, the comparison experiments between the proposed algorithm and several classical algorithms were carried out on the simulated and real datasets with gradual characteristics, verifying the effectiveness of the proposed algorithm in dealing with non-stationary data problems.

There are two main problems about pneumothorax detection in X-ray images. The pneumothorax usually overlaps with tissues such as ribs and clavicles in X-ray images, easily causing missed diagnosis and the performance of the existing pneumothorax detection methods remain to be improved. The suspicious pneumothorax area detection cannot be exploited by the convolutional neural network-based algorithms, lacking the interpretability. Aiming at the problems, a novel method combining Dense convolutional Network (DenseNet) and gradient-weighted class activation mapping was proposed. Firstly, a large-scale chest X-ray dataset named PX-ray was constructed for model training and testing. Secondly, the output node of the DenseNet was modified and a sigmoid function was added after the fully connected layer to classify the chest X-ray images. In the training process, the weight of cross entropy loss function was set to alleviate the problem of data imbalance and improve the accuracy of the model. Finally, the parameters of the last convolutional layer of the network and the corresponding gradients were extracted, and the areas of the pneumothorax type were roughly located by gradient-weighted class activation mapping. The experimental results show that, the proposed method has the detection accuracy of 95.45%, and has the indicators such as Area Under Curve (AUC), sensitivity, specificity all higher than 0.9, performs the classic algorithms of VGG19, GoogLeNet and ResNet, and realizes the visualization of pneumothorax area.

Existing optimizations have low precision and slow speed for reconfiguration of distribution network. In order to improve the safety and reliability of distribution network with Distributed Generation (DG), a simplified particle swarm optimization with adaptive inertial weight and full information was proposed based on leap-frog grouping. Firstly, from the viewpoints of reducing the active power loss of the network, increasing the voltage stability, and balancing the load of the feeder, a multi-objective mathematical model for distribution network was established. Secondly, through the Pareto dominance principle, the multi-objective was converted into several single objects with the same dimension, the same attribute and the same order of magnitude according to the standardized satisfaction of fuzzy membership function to make up for the disadvantages subjectivity and disunited dimension of weight method. Finally, in order to avoid random initialization to produce a large number of infeasible solutions, a kind of multi-objective reconfiguration strategy of distribution network with DG-combining Ant Colony Optimization (ACO) algorithm with random spanning tree and improved particle swarm optimization was designed. Through the IEEE33 node distribution system simulation, the experimental results show that the proposed reconfiguration strategy has a decrease of 41.0% in search efficiency compared to Particle Swarm Optimization (PSO) algorithm. Compared to before reconfiguration, the active power loss of the network is decreased by 41.47%, the voltage stability is decreased by 57.0%, and the load of the feeder is improved by 31.25%. The reconfiguration strategy effectively improves the optimizing accuracy and speeds up the optimization, therefore, improves the safety and reliability of distribution network operation.

There are some problems in big data that the chunking size of the deduplication algorithm for Content-Defined Chunking (CDC) is difficult to control, the expense of fingerprint calculation and comparison is high, and the parameter needs to be set in advance. Thus, a Deduplication algorithm based on Winnowing Fingerprint Matching (DWFM) was proposed. Firstly, the chunking size prediction model was introduced before chunking, which can accurately calculate proper chunking size according to the application scenario. Then, the ASCⅡ/Unicode was used as the data block fingerprint in the calculation of the fingerprint. Finally, when determining the block boundary, the proposed algorithm based on chunk fingerprint matching does not need to set the parameters in advance to reduce fingerprint calculation and contrast overhead. The experimental results on a variety of datasets show that DWFM is about 10% higher than FSP (Fixed-Sized Partitioning) and CDC algorithms in deduplication rate, and about 18% in fingerprint computing and contrast overhead. As a result, the chunking size and boundaries of DWFM are more consistent with data characteristics, reducing the impact of parameter settings on the performance of deduplication algorithms, meanwhile, effectively eliminating more duplicate data when dealing with different types of data.

The software development process is increasingly depending on various Computer-Aided Software (CAS). Simultaneously using these tools bring some problems, including non-customized development process, inconsistent process data and inefficient process management. To deal with these problems, a software development process management system that supports multi-tool collaboration was proposed. The hierarchical architecture system was developed on workflow design by analyzing software development process and studying the software engineering development model that supports fast iteration and tends to process management. Besides, the system was rigorously tested under 576 test cases. As a result the pass rate is 85%, which is able to meet the majority of tool collaboration needs, including definable development process, consistent interaction data and available process management. The system has been used by seven development teams with about 200 developers. The feedback results from the managers, developers and testers show that this system saves the time of weekly meetings, facilitates the management of development tasks, and significantly improves the development efficiency.

Aiming at the problems of massive data analysis requirement, complex cognitive inference in big data tasks, low efficiency of random assignment algorithm and virtual property and uncertainty of Internet users, an optimization algorithm for accurately theme-aware task assignment in crowd computing on big data was proposed. Firstly, the themes in crowd computing were extracted by method which combined with theme extraction model with fuzzy-kmeans adaptation, then the correlations were computed through task model and user model. Secondly, new users' real theme and initial accuracy were tested by historical tasks with high quality answers. Lastly, the probability that a user can participate in a certain kind of task was calculated and a sequence of candidate sequences was predicted by Logistic Regression (LR), and then the appropriate workers were assigned accurately to the tasks. Compared with random algorithm, the accuracy of the proposed algorithm was more than 20 percentage points higher, which increases with the increase of the training data, and the accuracy was nearly close to 100% especially in correlation tasks through full training. The simulation results show that the proposed algorithm has a higher accuracy with more cost-effective and performance in big data environment.

In order to solve the problems of keyword extraction and project keyword lexicon establishment of technological projects in professional fields, an algorithm for building the lexicon based on semantic relation and co-occurrence matrix was proposed. On the basis of conventional keyword extraction research based on co-occurrence matrix, the algorithm considered several advanced factors such as the location, property and Inverse Document Frequency (IDF) index of the keywords to improve the traditional approach. Meanwhile, a method was given for the establishment of keyword semantic network using co-occurrence matrix and hot keyword identification through computing the similarity with semantic base vector. At last, 882 project experiment documents in power field were used to perform the simulation. And the experimental results show that the proposed algorithm can effectively extract the keywords for the technological projects, establish the keyword correlation network, and has better performance in precision, recall rate and F1-score than the keyword extraction algorithm of Chinese text based on multi-feature fusion.

To resolve the irrationality in the definition and measurement of intuitionistic fuzzy entropy, a new axiomatic definition for intuitionistic fuzzy entropy was proposed, and a new measuring formula was structured. Firstly, the existing differences in research of axiomatic definition for intuitionistic fuzzy entropy were analyzed, its defects and insufficiency were also pointed out. Secondly, an improved axiomatic definition for intuitionistic fuzzy entropy and a calculation formula of intuitionistic fuzzy entropy were proposed. Finally, the new formula was compared with the existing formulas for intuitionistic fuzzy entropy by examples. The results of the example analysis show that, the proposed entropy formula can reflect better the uncertainty and fuzziness of intuitionistic fuzzy sets, and the capability to discriminate the uncertainty of intuitionistic fuzzy sets is stronger.

To solve the problem that current Decentralized Information Flow Control (DIFC) systems are unable to monitor the integration of host and network sensitive data effectively, a new design framework of DIFC system based on Software Defined Network (SDN), called S-DIFC, was proposed. Firstly, this framework used DIFC modules to monitor files and processes in host plane with fine granularity. Moreover, label mapping modules were used to block network communication and insert sensitive data labels into network flow. Meanwhile the multi-level access control of the flow with security label was implemented with SDN's controller in network plane. Finally, S-DIFC recovered security labels carried by sensitive data in DIFC system on target host. The experimental results show S-DIFC influences host with CPU performance decrease within 10% and memory performance decrease within 1.3%. Compared to Dstar system with extra time-delay more than 15 seconds, S-DIFC mitigates communication overhead of distributed network control system effectively. This framework can meet the sensitive data security requirements of next generation network. In addition, the distributed method can enhance the flexibility of monitor system.

Oblivious transfer plays an important role in the field of cryptography. A provably secure k-out-of-n oblivious transfer scheme was analyzed in this paper. This scheme was based on a novel method and was efficient in computation and communication. However, it was found not secure at all after deep analysis. The main fault is that the receiver can easily acquire all the secret messages sent by sender. Thus it does not satisfy the secure requirement of oblivious transfer. Finally, by adding a random number the fault of the scheme was fixed. The improved k-out-of-n oblivious transfer scheme keeps the same communicational overhead and computational overhead as the original one. The security of the improved scheme is also based on Decisional Diffie-Hellman (DDH) assumption.

Concerning the large-scale concurrent video stream scheduling problem of low resource utilization and load imbalance under cloud environment, a Video-on-Demand (VOD) scheduling policy based on Ant Colony Optimization (ACO) algorithm named VodAco was proposed. The correlation of video stream expected performance and server idle performance was analyzed, and a mathematical model was built based on the definition of comprehensive matching degree, then ACO method was adopted to hunt the best scheduling schemes. The contrast experiments with Round Robin (RR) and greedy schemes were tested on CloudSim. The experimental results show that the proposed policy has more obvious advantages in task completion time, platform resources occupancy and node load balancing performance.

The newly developed botnet defense technologies pose a severe challenge to botnet survivability. In order to improve the survivability of the botnet, from an attackers perspective, this article proposed a new anti-strike mechanism based on regroup, which was suitable for semi-distributed botnet. In the case that semi-distributed botnet suffered a severe blow, which caused topology broken, this mechanism could perceive the state of botnet, detect survival nodes, recover survival node and reassemble them into a new botnet. The experiments verify the effectiveness of the mechanism to effectively enhance the survivability of the semi-distributed botnet.

At present there are many kinds of technologies which can track, detect and counter botnet effectively, which are serious threats to botnet. In order to improve the survivability of botnets, with the analysis on the existing anti-botnet technology, the paper proposd a new P2P-botnet based on Kademlia protocol from an attacker's prospective. A communication encryption and node authentication mechanism was designed. The theoretical analysis shows that the mechanism can effectively address improper command attack and sybil attack. Eventually, the experimental results verify that this botnet has high survivability.

In order to achieve the high efficiency in automatic train operation, on the basis of the analysis of the train at different stages of operation, taking parking as the key stage, analytic hierarchy process was used to get quantitative description of the importance between each performance indexes and evaluation function of parking controls comprehensive performance in this stage, then the fuzzy manipulation rules of the online control were got. The offline operation of train under the rules was simulated for several times, the different schemes in sub-regional division and start braking point selection were scored to get the parking manipulation scheme which performance indexes are the best. Finally the simulation system was designed based on VC++ platform, and it has verified that the practical effect of the train running under the control algorithm has good parking precision, comfort and time saving.

In this paper, a new method based on Predictive State Representation (PSR) was proposed to solve high complexity of multivariate probability system. The paper introduced a new concept about general multivariate process in the first, and then described the multivariate system with the concept. Furthermore, the authors imported MultiVariate (MV)PSR as prediction model for multivariate system. The model was based on observable information and could realize the multivariate prediction in the finite dimensions. The experimental result shows that the approximate model effectively reduces the complexity of the system prediction.

Based on certain detection characteristics of process, hidden process could be uncovered by memory searching. However, malware, with the help of developing Rootkit, could hardly be detected because its feature has been manipulated or virtual memory scan could be invalid, thus increasing the difficulty of detection. In order to address this issue, a new multi-characteristics matching approach was proposed. It was to obtain the whole physical memory image by Page Table Entry (PTE) patching, to extract the key fields from process data structure and construct a template to improve the reliability of characteristics, and to introduce similarity for preventing the detection leakage. The results show that the new detection is effective in the hidden process searching.

Multiprocessor scheduling algorithm is the key in the embedded real-time systems. According to the multiprocessor features, a new dynamic parallel scheduling algorithm of real-time multiprocessor, named Split-Parallel (SPara), was proposed. The algorithm solved the problem that the previous algorithms, such as Myopic, EDPF, only judge by the deadline to schedule the tasks, and it was also developed by adding the restriction of the urgency and an effective method as the task with long execution time and tight deadline. Furthermore, the multiprocessor scheduling algorithm which combined the theory of high-level coloured time Petri net was analyzed by modeling, and according to the model, an example of SPara algothrim was simulated and tested. The experimental results show that SPara performances are much better than the other algorithms like Myopic in processor utilization and scheduling success ratio.

由于缺少结构化的表示,基于内容的图像分类存在一定的问题,据此提出了一种基于迭 代神经网络的自然图像表示和分类的方法。利用Berkeley分割算法将图像分割成不同的区域,采用 基于人工的多叉树或基于邻接区域的二叉树的方法进行区域合并,同时提取区域统计特征,得到图像 的树型结构表示。根据BPTS算法对网络进行训练,训练好的网络就具备了图像分类的功能。实验 结果表明,基于迭代神经网络的结构表示和分类方法具有很强的结构学习能力,同时人工生成的多叉 树涵盖更多的语义信息且能得到较好的分类结果。

This paper analyzed the principle of collecting information of network topology with SNMP and ICMP, the faults of these means and some advice were discussed. Then, a new RT algorithm of remote network topology by synthetizing many probing techniques was presented. The algorithm not only put emphasis on the collection of network devices infomation,but also gave prominence to the analysis of these infomation. Finaly, the integrality and veracity of the result of this new network topology discovery algorithm were proved by compared the results of different way,all based on an emulating network environment.