In order to solve the problems of high computing cost and over-fitting of the model caused by complicated network structure in pulmonary nodule detection method, an improved residual network structure combining deep separable convolution and pre-activation was proposed. And the proposed network structure was applied to a pulmonary nodule detection model. Based on the target detection network Faster R-CNN, with U-Net coder-decoder structure adopted, the deep separable convolution and pre-activation operations were used by the model to improve the 3D residual network structure. Firstly, with the use of deep separable convolution, the complexity and computing cost of the model were reduced. Then, the regularization of the model was improved by introducing the pre-activation operation, and the phenomenon of overfitting was alleviated. Finally, the rectangular convolution kernel was used to expand the receptive field of the convolution operation on the premise that the computing cost of the model was slightly increased, so as to effectively take into account both the global and local characteristics of the pulmonary nodules. On the LUNA16 dataset, the proposed method has the sensitivity of 96.04%, and the Free-response area under the Receiver Operating Characteristic curve (FROC) score of 83.23%. The experimental results show that the method improves the sensitivity of pulmonary nodule detection, effectively reduces the average number of false positives in the detection results, and improves the detection efficiency. This proposed method can effectively assist radiologists in detecting pulmonary nodules.

In order to enhance the network representation learning quality with node profile information, and focus on the problems of semantic information dispersion and incompleteness of node profile attribute information in social network, a network representation learning algorithm incorporated with node profile information was proposed, namely NPA-NRL. Firstly, attribute information were encoded by one-hot encoding, and a data augmentation method of random perturbation was introduced to overcome the incompleteness of node profile attribute information. Then, attribute coding and structure coding were combined as the input of deep neural network to realize mutual complementation of the two types of information. Finally, an attribute similarity measure function based on network homogeneity and a structural similarity measure function based on SkipGram model were designed to mine fused semantic information through joint training. The experimental results on three real network datasets including GPLUS, OKLAHOMA and UNC demonstrate that, compared with the classic DeepWalk, Text-Associated DeepWalk (TADW), User Profile Preserving Social Network Embedding (UPP-SNE) and Social Network Embedding (SNE) algorithms, the proposed NPA-NRL algorithm has a 2.75% improvement in average Area Under Curve of ROC (AUC) value on link prediction task, and a 7.10% improvement in average F1 value on node classification task.

In hyperspectral remote sensing image classification, the traditional active learning algorithms only use labeled data for training sample, massive unlabeled data is ignored. In order to solve the problem, a new active learning algorithm combined with unlabeled information was proposed. Firstly, by realizing triple screening of K neighbor consistency principle,predict consistency principle, and information evaluation of active learning, the unlabeled sample with a certain amount of information and highly reliable prediction label was obtained. Then, the prediction label was added to the label sample set as real label. Finally, an optimized classification model was produced by training the sample. The experimental results show that, compared with the passive learning algorithms and the traditional active learning algorithms, the proposed algorithm can obtain higher classification accuracy under the precondition of the same manual labeling cost and get better parameter sensitivity.

To realize hybrid Web service composition when Web Services Description Language (WSDL) and Web Ontology Language for Service (OWL-S) coexist, an approach based on interface automata was presented. Firstly, the interface automata were employed to accomplish automatic recognition and composition of Web services after analyzing the relations between WSDL and OWL-S. Simultaneously, the optimal results were obtained to realize different service business logic by comparing the service composition with the predefined service quality. The results of a tourism service sample show that the approach is feasible and effective, and the efficiency of service composition is improved by 5%-10%.

In the service computing mode, affection behaviour was employed to improve the efficiency of service composition. Firstly, an affection space was built to meet behaviour demands, and cognition was defined to reason state change of affection. In the change processing, mapping was done between affection and cognition, and emotion decay and emotion update were defined to maintain the stability of affective change. Secondly, affective mechanism was put into ant colony algorithm, which formed an affection ant colony algorithm, and the algorithm was applied to Web Service Modeling Ontology (WSMO) service composition. Finally, the paper adopted a Virtual Travel Agency (VTA) example under WSMO to show this approach was effective and feasible.

There are some SaaS problems for Web service and REST (Representational State Transfer) interfaces recognition in the software generation process. Therefore, an approach was proposed based on partitioning algorithm, which employed partitioning algorithm to implement function partition of SaaS and define difference nodes for difference functions. At the same time, the similarity between nodes was defined to accomplish partition, which improved the efficiency of SaaS functions. Secondly, according to the changing requirements, addressing and interrupt approach was presented to realize software generation of SaaS. Finally, an SaaS online sale software in Amazon cloud computing was analyzed, which approves that the approach is feasible and available.

A distributed and dynamic load balancing mechanism was proposed.To preserve per-flow packet ordering and support unequal weighted distribution,hashing-based algorithm with dynamic adapting according to history records on traffic was used.Simulation experiments evaluate the effects of hash-tuning with varied time window and threshold.

IETF put forward a new version of IKE, IKEv2.Different from the old IKE,IKEv2 combines and redefines key exchange process. This paper introduced IKEv2, and analysed the security of key negotiation mechanism of IKEv2.Aiming at some security problems in EAP exchange, such as authentication with digital certificate and reauthentication to avoid accessing VPN tunnel with unauthorized identity, some improvement advice and solutions were given.

Chaos is widely used in image encryption because of its high sensitivity to initial conditions and parameters and its stochastic series. A kind of chaos map whose parameters were randomly changed was brought forward. A chaos key stream with good randomcity and long cycle was made out, whose statistic character was proved strictly. The key stream has been used to encrypt image, and method was proved to work well.