As the graph-based query in SPARQL query processing becames more and more inefficient due to the increasing structure complexity of Resource Description Framework (RDF) in the graph, by analyzing the basic structure of RDF graphs and the selectivity of the RDF vertices, RDF Triple Patterns Selectivity (RTPS) was proposed to improve the efficienccy of subgraph matching for graph with RDF, which is a graph structure segmentation rule based on selectivity of RDF vertices. Firstly, according the commonality of the predicate structure in the data graph and the query graph, an RDF Adjacent Predicate Path (RAPP) index was built, and the data graph structure was transformed into incoming-outgoing predicate path structure to determine the search space of query vertices and speed up the filtering of RDF vertices. Secondly, the model of Integer Linear Programming (ILP) problem was built to divide a RDF query graph with complicated structure into several query subgraphs with simple structure. By analyzing the structure characteristics of the RDF vertices in the adjacent subgraphs, the selectivity of the query vertices was established and the optimal segmentation method was determined. Thirdly, with the searching space narrowed down by the RDF vertex selectivity and structure characteristics of adjacent subgraphs, the matchable RDF vertices in the data graph were found. Finally, the RDF data graph was traversed to find the subgraphs whose structure matched the structure of query subgraphs. Then, the result graph was output by joining the subgraphs together. The controlling variable method was used in the experiment to compare the query response time of RTPS, RDF Subgraph Matiching (RSM), RDF-3X, GraSS and R3F. The experimental results show that, compared with the other four methods, when the number of triple patterns in a query graph is more than 9, RTPS has shorter query response time and higher query efficiency.

With the complexity increasing of query graph structure, the efficiency of graph-based query in SPARQL query processing becomes lower and lower. By analyzing the basic structure of Resource Description Framework (RDF) graph, a subgraph matching method based on structure segmentation of query graph, called RSM (RDF Subgraph Matching), was proposed. Firstly, a query graph was divided into several simple query subgraphs, and query graph node searching space was defined through structure index of adjacent predicate. Secondly, the searching space was narrowed down by the adjacent subgraph structure, and a matchable subgraph could be found in data graph according to the searching area in the searching space. Finally, the result graph was output by joining related subgraphs. The query response times of RSM, RDF-3X, R3F and GraSS on query graphs with different structural complexity in different data sets were compared. The experimental results show that, compared with the other three methods, RSM has a shorter query response time and higher query efficiency in processing complex query graphs.

In the field of medical imaging analysis using machine learning, training samples are not enough. In order to solve the problem, a multimodal multi-label transfer learning model was proposed and applied to early diagnosis of Alzheimer's Disease (AD). Specifically, the multimodal multi-label transfer learning model consisted of two components:multi-label transfer learning feature selection and multimodal multi-label learning machine for classification and regression together. Firstly, the multi-label transfer learning feature selection model was built, which was based on the conventional sparse multi-label learning of Lasso (Least absolute shrinkage and selection operator) model for the combination of classification and regression tasks. Secondly, the technique of transfer learning was used to extend the conventional sparse multi-label learning of Lasso model and create the multi-label transfer learning feature selection model that can be performed on training samples from different learning multi-domains. Then, according to the multimodal feature data in the heterogeneous feature space, the multi-kernel learning was used to combine multimodal feature kernel matrix. Finally, the multimodal multi-label learning machine was built, and which was consisted of multi-kernel learning for the combination of multimodal biomarkers and multi-label classification and regression model. To evaluate the effectiveness of the multimodal multi-label transfer learning model, the Alzheimer's Disease Neuroimaging Initiative (ADNI) database was employed. The experimental results on the ADNI database show that the proposed model can recognize Mild Cognitive Impairment Converters (MCI-C) patients from MCI NonConverters (MCI-NC) ones with 79.1% accuracy and predict clinical scores with 0.727 correlation coefficient, so it can significantly improve the performance of early AD diagnosis with the aid of related domain knowledge.

String similarity join is an essential operation of data quality management and a key step to find the value of data. Now in the era of big data, since the existing methods can not meet the demands of incremental processing, an incremental string similarity join method oriented streaming data, called Inc-Join, was proposed. And the string index technique was optimized. Firstly, based on the Pass-Join string join algorithm, strings were divided into some disjoint substrings by utilizing partition technique; secondly, the inverted index of strings was created and acted as a state; finally, the similarity calculation was done according to the state when new data came, and the state would be updated after each operation of string similarity join. The experimental results show that Inc-Join method can reduce the number of reduplicate matching between short or long strings to √ n( n is the number of matching with batching processing model) without affecting the join accuracy. The elapsed time of string similarity join with batching processing model was 1 to 4.7 times the time Inc-Join needs when three different datasets were processed, and it tended to increase sharply. And the minimum elapsed time of optimized Inc-Join only accounted for 3/4 of original elapsed time of Inc-Join. With the increasing number of strings, the elapsed time of optimized Inc-Join would account for less and less of proportion in original elapsed time. The state need not to be saved, so the optimized Inc-Join further reduces time and space cost of Inc-Join.

Concerning the lackness of effective means in the feature evaluation of Advanced Radar Emitter Signals (ARES), and the excessive dependence on expert experience in Analytic Hierarchy Process (AHP), a new feature evaluation model of ARES named SPA-FAHP was proposed based on Set Pair Analysis (SPA) and Fuzzy Analytic Hierarchy Process (FAHP). In order to solve the uncertainty or fuzzy judgement of the judge people when they evaluate the large-capacity data of radar emitter signals, the traditional AHP was improved through the introduction of triangular fuzzy numbers, and the index weights of ARES feature evaluation system were analyzed by FAHP. Then, the expert decision matrix of traditional AHP was made improvement and identical degree analysis through the introduction of SPA theory to solve the problem that the decisions of AHP rely on experience of experts too much. Finally, ARES features were made comprehensive evaluation through the combination of index weights matrix and identical degree matrix of the decision. The calculation results show that the model is effective and feasible. It can achieve the characteristic analysis and evaluation of ARES features more objectively.

Concerning the slow convergence of distributed power control algorithm in cognitive radio, a novel algorithm based on non-cooperative game was proposed for cognitive radio system. A Signal-to-Interference Ratio (SIR)-based tangent cost function of fewer iterations was designed to improve the convergence. The simulation results demonstrate that, compared with the Koskie-Gajic algorithm and the Cognitive Radios-Non-Cooperative Power Control Game (CR-NCPCG) algorithm, the proposed algorithm at the premise of satisfying the secondary users' demand for SIR and the primary user's interference temperature constraint, not only can fast converge, but also has higher average SIR with at least 0.3dB when the number of users is less than twenty. It can control the power of secondary users effectively.

Against the massive monitoring data of lightning locating system, a lightning nowcasting model based on Improved Density-Based Spatial Clustering of Application with Noise (IDBSCAN) clustering algorithm was put forward. Based on the lightning location data in real-time monitoring system, this method searched for lightning-density flash point greater than the threshold value of the land, built the cluster with up to the maximum ground flash density, and located the core of the cluster. Besides, with the application of adjacency list search algorithm, time and space consumed for the initial search set of lightning data had been greatly reduced. Furthermore, using regression fitting algorithm, the proposed algorithm can predict the path of movement of lightning cluster. The experimental results show that IDBSCAN algorithm used in the lightning nowcasting is effective.

In allusion to the insufficiency of current P2P Geographic Information System (GIS) in utilizing network resources of clients, based on analyzing and summarizing the existing hybrid P2P spatial indexing network, a new idea of group strategy was proposed in view of practice. In this idea, peers with the same spatial data semantics were joined in the same group in which the burden of query was shared by group members together. Furthermore, a replacement algorithm of current index nodes and backup strategy were proposed to improve the query performance and stability of the overall network. The experimental results indicate that the indexing network with group strategy can effectively make use of clients' network resources and improve the query performance when a large number of queries request concurrently.

The multicast tree in MAODV is reconstructed frequently because of the nodes mobility, making the cost of routing and delay of transmission increase significantly. The neighbor change ratio based stable path selection method was proposed to overcome the shortcomings. And a new neighbor change ratio calculation method was put forward, which did not need to send Hello messages timely. Based on this new method, a stable path based MAODV (SP-MAODV) multicast routing protocol was given with a stable path selection and less hops. The simulation results on data packet transmission rate, routing overhead, average end-to-end delay and delay jitter show that the new protocol reduces interruption probability of the path and improves performance of the protocol.