Graph neural networks (GNNs) have received much attention as GNNs have recently been successfully applied on non-euclidean data. However, artificially designed graph neural networks often fail to get satisfactory model performance for a given graph data. Graph neural architecture search effectively constructs the GNNs that achieve the expected model performance with the rise of automatic machine learning. The challenge is efficiently and automatically getting the optimal GNN architecture in a vast search space. Existing search methods serially evaluate the GNN architectures, severely limiting system efficiency. To solve these problems, we develop an Auto matic G raph N eural A rchitecture S earch framework (Auto-GNAS) with parallel estimation to implement an automatic graph neural search process that requires almost no manual intervention. In Auto-GNAS, we design the search algorithm with multiple genetic searchers. Each searcher can simultaneously use evaluation feedback information, information entropy, and search results from other searchers based on sharing mechanism to improve the search efficiency. As far as we know, this is the first work using parallel computing to improve the system efficiency of graph neural architecture search. According to the experiment on the real datasets, Auto-GNAS obtain competitive model performance and better search efficiency than other search algorithms. Since the parallel estimation ability of Auto-GNAS is independent of search algorithms, we expand different search algorithms based on Auto-GNAS for scalability experiments. The results show that Auto-GNAS with varying search algorithms can achieve nearly linear acceleration with the increase of computing resources.

Studying the similarity of diseases can help us to explore the pathological characteristics of complex diseases, and help provide reliable reference information for inferring the relationship between new diseases and known diseases, so as to develop effective treatment plans. To obtain the similarity of the disease, most previous methods either use a single similarity metric such as semantic score, functional score from single data source, or utilize weighting coefficients to simply combine multiple metrics with different dimensions. In this paper, we proposes a method to predict the similarity of diseases by node representation learning. We first integrate the semantic score and topological score between diseases by combining multiple data sources. Then for each disease, its integrated scores with all other diseases are utilized to map it into a vector of the same spatial dimension, and the vectors are used to measure and comprehensively analyze the similarity between diseases. Lastly, we conduct comparative experiment based on benchmark set and other disease nodes outside the benchmark set. Using the statistics such as average, variance, and coefficient of variation in the benchmark set to evaluate multiple methods demonstrates the effectiveness of our approach in the prediction of similar diseases.

Power knowledge graph (PKG) plays an essential role in the electricity power field to integrate multi-source heterogeneous power data in a structural form. Recently, PKG is widely used for management and analysis of electric power big data. However, a single PKG can only provide limited power knowledge, which restricted its functions in downstream applications. In this paper, we merge multiple power knowledge graphs by aligning the entities which belong to different sources or languages respectively. To align entities, a Typed Graph Attention Network method is proposed to learn neighbor-level embedding through the weight of relation types in power knowledge graph. By incorporating structural information and relation type information simultaneously, the proposed method can accurately capture entity features in different Power knowledge graph.

Interactive volume data visualization using a mixed reality (MR) system is increasingly popular in the research and development field of computer graphics. Due to sophisticated requirements of user interaction and vision when using head-mounted display (HMD) devices, the conflict between the realisticness and efficiency of direct volume rendering (DVR) is yet to be resolved for MR applications. We develop an MR visualization framework that can support inter-active realistic volume rendering and integrated it into a prototyping MR system. This MR system can display high-quality volume ren-dering effects that synchronously reflect dyanamic illumination of surrounding environment in real time. The comparative analysis shows that our MR visualization framework helps provide users intu-itive perception of displayed volumetric structures as they naturally exist in the real-world environment.