Aiming at the problem that the imbalance of datasets in credit risk assessment affects the prediction effect of the model， a credit risk prediction model based on Borderline Adaptive Synthetic Minority Oversampling TEchnique （BA-SMOTE） and Focal Loss-Light Gradient Boosting Machine （FLLightGBM） was proposed. Firstly， on the basis of Borderline Synthetic Minority Oversampling TEchnique （Borderline-SMOTE）， the adaptive idea and new interpolation method were introduced， so that different numbers of new samples were generated for each minority sample at the border， and the positions of the new samples were closer to the original minority sample， thereby balancing the dataset. Secondly， the Focal Loss function was used to improve the loss function of LightGBM （Light Gradient Boosting Machine） algorithm， and the improved algorithm was used to train a new dataset to obtain the final BA-SMOTE-FLLightGBM model constructed by BA-SMOTE method and FLLightGBM algorithm. Finally， on Lending Club dataset， the credit risk prediction was performed. Experimental results show that compared with other imbalanced classification algorithms RUSBoost （Random Under-Sampling with adaBoost）， CUSBoost （Cluster-based Under-Sampling with adaBoost）， KSMOTE-AdaBoost （K-means clustering SMOTE with AdaBoost）， and AK-SMOTE-Catboost （AllKnn-SMOTE-Catboost）， the constructed model has a significant improvement on two evaluation indicators G-mean and AUC （Area Under Curve） with 9.0%-31.3% and 5.0%-14.1% respectively. The above results verify that the proposed model has a better default prediction effect in credit risk assessment.

To improve the time and space efficiency of Frequent Pattern (FP) mining algorithm over uncertain dataset, the Uncertain Frequent Pattern Mining based on Max Probability (UFPM-MP) algorithm was proposed. First, the expected support number was estimated using maximum probability of the transaction itemset. Second, by comparing this expected support number to the minimum expected support number threshold, the candidate frequent itemsets were identified. Finally, the corresponding sub-trees were built for recursively mining frequent patterns. The UFPM-MP algorithm was tested on 6 classical datasets against the state-of-the-art algorithm AT (Array based tail node Tree structure)-Mine with positive results (about 30% improvement for sparse datasets, and 3-4 times more efficient for dense datasets). The expected support number estimation strategy effectively reduces the number of sub-trees and items of header table, and improves the algorithm's time and space efficiency; and when the minimum expected support threshold is low or there are lots of potential frequent patterns, time efficiency of the proposed algorithm performs more remarkably.

The conventional Receiver Autonomous Integrity Monitoring (RAIM) algorithm is limited when detecting weak pseudo-range bias under gradual change because of its longer detection delay and higher miss detection rate. A weight modification accumulated epochs parity vector RAIM algorithm based on self-adaptive strategy was presented to solve this problem. In this algorithm, the weight factor was obtained according to the single epoch fault degree to adjust the proportion of each epoch in the selected window to structure more effective detection statistics, and the size of the window was determined according to the repeated simulation experiments. The simulation results show that the proposed method can better detect weak pseudo-range bias under gradual change, compared to accumulated epoch and the conventional RAIM algorithm, the detection delay time declines by 16.67% and 56.52% respectively.

Concerning the limitations of the existing clustering methods based on intuitionistic fuzzy sets, a clustering model called Weighted Intuitionistic Fuzzy Set Model (WIFSCM) was proposed based on weighted intuitionistic fuzzy sets. In this model, the concepts of equivalent sample and weighted intuitionistic fuzzy set were put forward in special feature space, and based on which the objective function of intuitionistic fuzzy clustering algorithm was proposed. Iterative algorithms of clustering center and matrix of membership degree were inferred from the objective function. The density function based on weighted intuitionistic fuzzy sets was defined, and initial clustering center was gotten to reduce iterative times. The experiment of gray image segmentation shows that WIFSCM is effective, and it is faster than IFCM algorithm nearly a hundred times.

In order to improve Windows Media Audio (WMA) decoder performance and obtain its exact performance, a new Standalone development platform was proposed. The critical step of Standalone platform is to configure i.MX31 Memory Management Unit (MMU). The segment translation mechanism was used to implement MMU. Via analysis of WMA decoder experiment results on this platform, WMA decoder exact performance was achieved, and its performance was enhanced by twenty-nine percent.