To address the problem of bandwidth resource management in Web applications, a prediction method for bandwidth requirement and Quality of Service (QoS) of Web applications based on network simulation was proposed. A modeling framework and formal specification were presented for Web services, a simplified parallel workload model was adopted, the model parameters were extracted from Web application access logs by means of automated data mining, and the complex network transmission process was simulated by using network simulation tool. As a result, the bandwidth requirement and changes on QoS were able to be predicted under different workload intensities. A classic benchmark system named TPC-W was used to evaluate the accuracy of prediction results. Theoretical analysis and simulation results show that compared with traditional linear regression prediction, network simulation can stably simulate real system, the predicted average relative error for total request number and total byte number is 4.6% and 3.3% respectively. Finally, with different bandwidth scaling schemes simulated and evaluated based on the TPC-W benchmark system, the results can provide decision support for resource management of Web applications.

Most existing cloud storage systems are based on the model, which leads to a full dataset scan for multi-dimensional queries and low query efficiency. A KD-tree and R-tree based multi-dimensional cloud data index named KD-R index was proposed. KD-R index adopted two-layer architecture: a KD-tree based global index was built in the global server and R-tree based local indexes were built in local server. A cost model was used to adaptively select appropriate R-tree nodes to publish into global KD-tree index. The experimental results show that, compared with R-tree based global index, KD-R index is efficient for multi-dimensional range queries, and it has high availability in the case of server failure.

To improve the quality of video surveillance outdoors and to automatically acquire the weather situations, a method to recognize weather situations in outdoor images is presented. It extracted such parameters as power spectrum slope, contrast, noise, saturation as features to realize the multi-classification of weather situations with Support Vector Machine (SVM). Then a decision tree was constructed in accordance with the distance between these features. The experimental results on WILD image base and our image set of eight hundred samples show that the proposed method can recognize sunny, overcast, foggy weather more than 85%, and recognize rainy weather more than 75%.

A novel algorithm for the creation of classifier ensemble based on fuzzy clustering was introduced. The algorithm got the distribution characteristics of the training sets by fuzzy clustering and sampled different training dataset to train different individual classifiers. Then the algorithm adjusted every sample's weight to get more classifiers through evaluating the quality of the classifier until certain termination condition was satisfied. The algorithm was tested on the UCI benchmark data sets and compared with two other classical algorithms: AdaBoost and Bagging. Results show that the new algorithm is more robust and has higher accuracy.