Concerning the current lack of effective development and deployment tools for deep learning applications， a component-based development framework for deep learning applications was proposed. The framework splits functions according to the type of resource consumption， uses a review-guided resource allocation scheme for bottleneck elimination， and uses a step-by-step boxing scheme for function placement that takes into account high CPU utilization and low memory overhead. The real-time license plate number detection application developed based on this framework achieved 82% GPU utilization in throughput-first mode， 0.73 s average application latency in latency-first mode， and 68.8% average CPU utilization in three modes （throughput-first mode， latency-first mode， and balanced throughput/latency mode）. The experimental results show that based on this framework， a balanced configuration of hardware throughput and application latency can be performed to efficiently utilize the computing resources of the platform in the throughput-first mode and meet the low latency requirements of the applications in the latency-first mode. Compared with MediaPipe， the use of this framework enabled ultra-real-time multi-person pose estimation application development， and the detection frame rate of the application was improved by up to 1 077%. The experimental results show that the framework is an effective solution for deep learning application development and deployment on CPU-GPU heterogeneous servers.

Accurate prediction of taxi demands between urban regions can provide decision support information for taxi guidance and scheduling as well as passenger travel recommendation， so as to optimize the relation between taxi supply and demand. However， most of the existing models only focus on modeling and predicting the taxi demands within a region， do not consider enough the spatial-temporal correlation between regions， and pay less attention to the more fine-grained demand prediction between regions. To solve the above problems， a prediction model for taxi demands between urban regions — Origin-Destination fusion with Spatial-Temporal Network （ODSTN） model was proposed. In this model， complex spatial-temporal correlations between regions was captured from spatial dimensions of the regions and region pairs respectively and three temporal dimensions of recent， daily and weekly periods by using graph convolution and attention mechanism， and a new path perception fusion mechanism was designed to combine the multi-angle features and finally realize the taxi demand prediction between urban regions. Experiments were carried out on two real taxi order datasets in Chengdu and Manhattan. The results show that the Mean Absolute Error （MAE）， Root Mean Square Error （RMSE） and Mean Absolute Percentage Error （MAPE） of ODSTN model are 0.897 1， 3.527 4， 50.655 6% and 0.589 6， 1.163 8， 61.079 4%， respectively， indicating that ODSTN model has high accuracy in taxi demand prediction tasks.

In order to improve the explainability of Recommendation System （RS）， break the inherent limitations of recommendation system and enhance the user’s trust and satisfaction on recommender systems， a Deep Review Attention Neural Network （DRANN） model with enhanced explainability was proposed. Based on the potential relationships between users and items on text reviews， the rich semantic information in user reviews and item reviews was used to predict users’ interest preferences and sentiment tendencies by the proposed model. Firstly， a Text Convolutional Neural Network （TextCNN） was used to do shallow feature extraction for word vectors. Then， the attention mechanism was used to assign weights to comment data and filter invalid comment information. At the same time， the deep autoencoder module was constructed to reduce the dimension of high-dimensional sparse data， remove interference information， learn deep semantic representation， and enhance the explainability of recommendation model. Finally， the prediction score was obtained through the prediction layer. Experimental results on the four public data sets including Patio， Automotive， Musical Instrument （M?I） and Beauty show that DRANN model has the smallest Root Mean Square Error （RMSE） compared with Probabilistic Matrix Factorization （PMF）， Single Value Decomposition++ （SVD++）， Deep Cooperative Neural Network （DeepCoNN）， Tree-enhanced Embedding Model （TEM）， DeepCF （Deep Collaborative Filtering） and DER（Dynamic Explainable Recommender）， verifying its effectiveness in improving performance and the feasibility of the adopted explanation strategy.

In integrated support engineering, the number of components in reliability block diagram is large, the level of mastering the principle of system is required to be high and the operational data is always incomplete. To resolve these problems, a method that identifies the reliability structure of system using the information of operational data and the reliability of the units was proposed. The system reliability was estimated by using the system performance information. In addition, all reliability structure models was traversed and the theoretical reliability was calculated with the system's units reliability information, then the deviations between the estimated value of system reliability and all the reliability theoretical values were calculated, and the identification results by the first N reliability structure models of the lowest deviation was outputted after sorting the deviations. The calculation results of a given example show that the combined system based on the voting reliability structure can be identified with the probability of around 80%, decreases to 3% of the scope out of all possible forms, it can significantly reduce the workload of the researcher to identify the system reliability structure.

At present, the safety analysis of SNAKE algorithm is mainly about interpolation attack and impossible differential attack. The paper evaluated the security of SNAKE(2) block cipher against integral attack. Based on the idea of higher-order integral attack, an 8-round distinguisher was designed. Using the distinguisher, integral attacks were made on 9/10 round SNAKE(2) block cipher. The attack results show that the 10-round SNAKE(2) block cipher is not immune to integral attack.

As far as the issue on Web user access pattern is concerned, adopting Maximal Frequent Path (MFP) can mine more universal patterns. A new user access pattern tree named WUAP-tree was devised. Furthermore, a new algorithm named WUAP-mine was proposed for mining user access patterns, which was based on E-OEM model for page topological structure and users' browse path. The algorithm utilized WUAP-tree that could neither generate candidate sets nor use recursive ways. It could mine frequent Web users access patterns by scanning transaction database and output-depth-first traversing WUAPtree only once. The algorithm is very easy to query Web user access patterns from WUAPtree directly. At last, theoretical analysis and experimental results prove its effectiveness and efficiency.