Emotion analysis can quickly and accurately dig out users’ emotional tendencies， and has a huge application market. Aiming at the complexity and diversity of the microblog language’s syntactic structures， a Syntax Graph Convolution Network （SGCN） model was proposed for fine-grained emotion classification of Chinese microblog. The proposed model has the characteristics of rich structural and semantic expression at the same time. In the model， a text graph was constructed on the basis of the dependency between words， and the correlation degree between words was quantified by Pointwise Mutual Information （PMI）. After that， the PMI was used as the weight of the corresponding edge to represent the structural information of the sentence. The semantic features fusing location information were taken as the initial features of nodes to increase the semantic features of nodes in the text graph. Experimental results on the microblog emotion classification dataset of Social Media Processing 2020 （SMP2020） show that for two sets of microblog data containing six categories of emotions： happiness， sadness， anger， fear， surprise， and emotionlessness， the average F1-score of the proposed model reaches 72.64% which is 2.75 and 3.87 percentage points higher than those of the BERT （Bidirectional Encoder Representations from Transformers） Graph Convolutional Network （BGCN） model and the Text Level Graph Neural Network （Text-Level-GNN） model， verifying that the proposed model can use the structural information of sentences more effectively to improve the classification performance than other deep learning models.

In order to improve the credibility of the damage degree evaluation to the aviation network due to cascading failures caused by emergency， considering the redundancy ability of airport nodes for the load， which means if the overload occurs in a certain spatial range， the node will not fail immediately but has a certain overload handling ability， an aviation network cascading failure model was proposed based on overload condition and failure probability. Firstly， the overload coefficient， weight coefficient， distribution coefficient， and capacity coefficient were introduced into the traditional "load-capacity" Motter-Lai cascading failure model. Then， the redundant capacity characteristics of network nodes were described by overload condition and failure probability， and different load redistribution strategies were applied to the failed and overloaded nodes to make the model more consistent with the aviation network reality. Theoretical analysis and simulation results show that increasing the overload coefficient within a certain range helps to reduce the impact of cascading failures， but the improvement effect is not obvious after increasing to a certain degree； with the optimal intervals for parameters of the model. the aviation network can maintain better robustness while spending smaller construction cost， and the optimized allocation of aviation network resources can improve the network’s resistance to cascading failures.

The model learned by ranking algorithm based on traditional pairwise method does not work well by ranking measure, such as Normalized Discounted Cumulative Gain (NDCG). To solve this problem, a novel ranking algorithm is proposed. The algorithm uses the same train data as the traditional way, what different is defining a new object function faced to NDCG. For the problem that the function is non-smooth, difficult to directly optimize, the algorithm presents to use the cutting plane algorithm which not only solve the problem above but also make the number of iteration not depending on the training size. Experimental results on the benchmark datasets prove the effectiveness of the proposed algorithm.