Aiming at the common non-linear relationship between characters in languages， in order to capture richer semantic features， a Named Entity Recognition （NER） method based on Graph Convolutional Network （GCN） and self-attention mechanism was proposed. Firstly， with the help of the effective extraction ability of character features of deep learning methods， the GCN was used to learn the global semantic features between characters， and the Bidirectional Long Short-Term Memory network （BiLSTM） was used to extract the context-dependent features of the characters. Secondly， the above features were fused and their internal importance was calculated by introducing a self-attention mechanism. Finally， the Conditional Random Field （CRF） was used to decode the optimal coding sequence from the fused features， which was used as the result of entity recognition. Experimental results show that compared with the method that only uses BiLSTM or CRF， the proposed method has the recognition precision increased by 2.39% and 15.2% respectively on MicroSoft Research Asia （MSRA） dataset and Biomedical Natural Language Processing/Natural Language Processing in Biomedical Applications （BioNLP/NLPBA） 2004 dataset， indicating that this method has good sequence labeling capability on both Chinese and English datasets， and has strong generalization capability.

In recent years， deep learning has been widely used in many fields. However， due to the highly nonlinear operation of deep neural network models， the interpretability of these models is poor， these models are often referred to as “black box” models， and cannot be applied to some key fields with high performance requirements. Therefore， it is very necessary to study the interpretability of deep learning. Firstly， deep learning was introduced briefly. Then， around the interpretability of deep learning， the existing research work was analyzed from eight aspects， including hidden layer visualization， Class Activation Mapping （CAM）， sensitivity analysis， frequency principle， robust disturbance test， information theory， interpretable module and optimization method. At the same time， the applications of deep learning in the fields of network security， recommender system， medical and social networks were demonstrated. Finally， the existing problems and future development directions of deep learning interpretability research were discussed.

In some industrial fields such as garments, shoes and metal building, it is often required to interpolate multiple specified set of skeleton curves using one loft developable surfaces (which can be developed onto plane without any distortion, tear or stretch). To solve this problem, one new algorithm based on Dijkstras algorithm was proposed to support the design of developable mesh surface: given multiple parametric curves, after adaptively sampling them, the objective is to search one globally optimal developable loft surface (given other specified objective surface energy function, varied corresponding loft mesh surfaces can be obtained of globally optimal topology). The problem was simplified into source route finding problem in Direct Acyclic Graph (DAG). The proposed algorithm is of practical engineering senses in designing loft surfaces in related engineering applications.